{"ep": "h2-00040", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "forbidden", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 29.3, "calls": 1, "tokens_in": 1964, "tokens_out": 2207, "ctx_chars": 0, "t_end": 1791533678.6238859, "error": "static check failed: line 87: getattr needs a literal, non-private attribute name; line 88: getattr needs a literal, non-private attribute name; line 89: getatt", "f_status": "forbidden", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 0.01, "compile_s": null, "r_error": "static check failed: line 87: getattr needs a literal, non-private attribute name; line 88: getattr needs a literal, non-private attribute name; line 89: getatt", "src_len": 3792, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1964, "tout": 2207, "ts": 1791533678.3683715, "chk": null, "code_len": 3792}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1964, "started": 1791533649.2849398}
{"ep": "h2-00036", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0071, "speedup": 1.0071, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 32.1, "calls": 1, "tokens_in": 6367, "tokens_out": 544, "ctx_chars": 10847, "t_end": 1791533681.3513434, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0071, "lo95": 1.0062, "mean": 1.018, "ev_only": 1.007, "ref_ms": 8.02892, "cand_ms": 7.97193, "n_trials": 160, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 25.31, "compile_s": 2.0, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 544, "ts": 1791533655.9437938, "chk": null, "code_len": 999}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791533649.281986}
{"ep": "h2-00045", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0065, "speedup": 1.0065, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 38.2, "calls": 1, "tokens_in": 6367, "tokens_out": 540, "ctx_chars": 10847, "t_end": 1791533687.4797099, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0065, "lo95": 1.0062, "mean": 1.0239, "ev_only": 1.007, "ref_ms": 7.98973, "cand_ms": 7.93848, "n_trials": 163, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 31.71, "compile_s": 1.85, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 540, "ts": 1791533655.691818, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791533649.2863352}
{"ep": "h2-00049", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7712, "speedup": 0.7712, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 52.4, "calls": 3, "tokens_in": 19133, "tokens_out": 1896, "ctx_chars": 8224, "t_end": 1791533701.6443832, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7712, "lo95": 0.7688, "mean": 0.7638, "ev_only": 0.7709, "ref_ms": 60.1855, "cand_ms": 78.03739, "n_trials": 19, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 15.39, "compile_s": 1.98, "r_error": "", "src_len": 1079, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 939, "ts": 1791533661.8163466, "chk": "ok", "code_len": 1079}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6576, "tout": 429, "ts": 1791533680.9911952, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7029, "tout": 528, "ts": 1791533685.985417, "chk": null, "code_len": 1079}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791533649.2898924}
{"ep": "h2-00016", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0539, "speedup": 1.0539, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 54.6, "calls": 3, "tokens_in": 22602, "tokens_out": 2236, "ctx_chars": 10527, "t_end": 1791533703.9076455, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0539, "lo95": 1.0532, "mean": 1.0934, "ev_only": 1.0542, "ref_ms": 19.60288, "cand_ms": 18.60036, "n_trials": 60, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0124, "r_wall_s": 18.69, "compile_s": 2.18, "r_error": "", "src_len": 2051, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1179, "ts": 1791533661.6172683, "chk": "ok", "code_len": 2051}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7952, "tout": 13, "ts": 1791533676.4546344, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7989, "tout": 1044, "ts": 1791533685.182541, "chk": null, "code_len": 2051}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791533649.274726}
{"ep": "h2-00001", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0059, "speedup": 1.0059, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 59.9, "calls": 2, "tokens_in": 4021, "tokens_out": 2043, "ctx_chars": 0, "t_end": 1791533709.1684303, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0059, "lo95": 1.0056, "mean": 1.0057, "ev_only": 1.0065, "ref_ms": 7.98395, "cand_ms": 7.93744, "n_trials": 135, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 23.71, "compile_s": 2.63, "r_error": "", "src_len": 1009, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1427, "ts": 1791533662.8457818, "chk": "ok", "code_len": 1009}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2518, "tout": 616, "ts": 1791533685.3029842, "chk": null, "code_len": 1009}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791533649.2658725}
{"ep": "h2-00046", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0037, "speedup": 1.0037, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 60.8, "calls": 3, "tokens_in": 23795, "tokens_out": 2581, "ctx_chars": 11823, "t_end": 1791533710.135085, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0037, "lo95": 1.0031, "mean": 1.0283, "ev_only": 1.0048, "ref_ms": 27.51024, "cand_ms": 27.4086, "n_trials": 39, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0226, "r_wall_s": 16.27, "compile_s": 1.95, "r_error": "", "src_len": 2267, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1344, "ts": 1791533663.266496, "chk": "ok", "code_len": 2374}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8377, "tout": 96, "ts": 1791533684.231509, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8497, "tout": 1141, "ts": 1791533693.822293, "chk": null, "code_len": 2267}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6921, "started": 1791533649.2874072}
{"ep": "h2-00056", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0212, "speedup": 1.0212, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 61.9, "calls": 2, "tokens_in": 4617, "tokens_out": 2280, "ctx_chars": 0, "t_end": 1791533711.1532133, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0212, "lo95": 1.0206, "mean": 1.0452, "ev_only": 1.0217, "ref_ms": 8.22504, "cand_ms": 8.05441, "n_trials": 184, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0083, "r_wall_s": 28.3, "compile_s": 2.07, "r_error": "", "src_len": 1329, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 1431, "ts": 1791533664.4144948, "chk": "ok", "code_len": 1329}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2865, "tout": 849, "ts": 1791533682.7209315, "chk": null, "code_len": 1329}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1752, "started": 1791533649.2900405}
{"ep": "h2-00034", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 63.3, "calls": 1, "tokens_in": 1723, "tokens_out": 978, "ctx_chars": 0, "t_end": 1791533712.5358968, "error": "RuntimeError: Error building extension 'fuse_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fuse_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval-p", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 52.78, "compile_s": 52.76, "r_error": "RuntimeError: Error building extension 'fuse_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fuse_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval-p", "src_len": 2209, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1723, "tout": 978, "ts": 1791533659.7110703, "chk": null, "code_len": 2209}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1723, "started": 1791533649.281168}
{"ep": "h2-00013", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.6433, "speedup": 0.6433, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 64.5, "calls": 2, "tokens_in": 15352, "tokens_out": 3261, "ctx_chars": 11065, "t_end": 1791533713.8201044, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.6433, "lo95": 0.6424, "mean": 0.6503, "ev_only": 0.6473, "ref_ms": 1.79123, "cand_ms": 2.7846, "n_trials": 195, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.5622, "r_wall_s": 13.47, "compile_s": 2.01, "r_error": "", "src_len": 2512, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 1986, "ts": 1791533672.5641613, "chk": "ok", "code_len": 2512}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8725, "tout": 1275, "ts": 1791533700.0959299, "chk": null, "code_len": 2512}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791533649.2725518}
{"ep": "h2-00030", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8852, "speedup": 0.8852, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 65.7, "calls": 3, "tokens_in": 18997, "tokens_out": 1679, "ctx_chars": 8224, "t_end": 1791533714.9944487, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8852, "lo95": 0.7183, "mean": 0.8534, "ev_only": 0.8852, "ref_ms": 67.89083, "cand_ms": 76.69375, "n_trials": 18, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 29.79, "compile_s": 2.45, "r_error": "", "src_len": 994, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 977, "ts": 1791533662.1635106, "chk": "ok", "code_len": 994}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6614, "tout": 217, "ts": 1791533680.5633829, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6855, "tout": 485, "ts": 1791533685.1458962, "chk": null, "code_len": 994}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791533649.2801208}
{"ep": "h2-00022", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7916, "speedup": 0.7916, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 66.5, "calls": 4, "tokens_in": 29576, "tokens_out": 1463, "ctx_chars": 10850, "t_end": 1791533715.812234, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.7916, "lo95": 0.7618, "mean": 0.7645, "ev_only": 0.787, "ref_ms": 0.17201, "cand_ms": 0.2173, "n_trials": 237, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 27.49, "compile_s": 2.96, "r_error": "", "src_len": 795, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 496, "ts": 1791533656.0143747, "chk": "ok", "code_len": 834}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 7227, "tout": 481, "ts": 1791533672.1006513, "chk": "ok", "code_len": 827}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7817, "tout": 69, "ts": 1791533684.060142, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7910, "tout": 417, "ts": 1791533688.178165, "chk": null, "code_len": 795}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 6622, "started": 1791533649.276914}
{"ep": "h2-00015", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0036, "speedup": 1.0036, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 67.0, "calls": 2, "tokens_in": 17298, "tokens_out": 4566, "ctx_chars": 11823, "t_end": 1791533716.2444172, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0036, "lo95": 1.0025, "mean": 1.0038, "ev_only": 1.004, "ref_ms": 27.54482, "cand_ms": 27.4449, "n_trials": 43, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0233, "r_wall_s": 14.84, "compile_s": 2.07, "r_error": "", "src_len": 2403, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 3344, "ts": 1791533679.8245454, "chk": "ok", "code_len": 2403}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10377, "tout": 1222, "ts": 1791533701.1473868, "chk": null, "code_len": 2403}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6921, "started": 1791533649.2742052}
{"ep": "h2-00007", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0227, "speedup": 1.0227, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 80.7, "calls": 3, "tokens_in": 15465, "tokens_out": 2441, "ctx_chars": 5026, "t_end": 1791533729.9671552, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0227, "lo95": 1.0217, "mean": 1.0222, "ev_only": 1.0221, "ref_ms": 8.1866, "cand_ms": 8.00518, "n_trials": 130, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0092, "r_wall_s": 39.47, "compile_s": 2.01, "r_error": "", "src_len": 1537, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1575, "ts": 1791533665.2628012, "chk": "ok", "code_len": 1537}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5680, "tout": 88, "ts": 1791533683.5679777, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5792, "tout": 778, "ts": 1791533690.4149861, "chk": null, "code_len": 1537}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 3993, "started": 1791533649.2691846}
{"ep": "h2-00062", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9989, "speedup": 0.9989, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 80.7, "calls": 2, "tokens_in": 13705, "tokens_out": 1179, "ctx_chars": 10909, "t_end": 1791533729.9847534, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9989, "lo95": 0.9979, "mean": 0.9991, "ev_only": 1.0014, "ref_ms": 8.02074, "cand_ms": 8.0297, "n_trials": 150, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 55.38, "compile_s": 1.91, "r_error": "", "src_len": 938, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 638, "ts": 1791533656.4849432, "chk": "ok", "code_len": 938}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7226, "tout": 541, "ts": 1791533674.536747, "chk": null, "code_len": 938}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791533649.2992496}
{"ep": "h2-00051", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8558, "speedup": 0.8558, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 80.9, "calls": 3, "tokens_in": 19089, "tokens_out": 1754, "ctx_chars": 8224, "t_end": 1791533730.1453485, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.8558, "lo95": 0.668, "mean": 0.8282, "ev_only": 0.8556, "ref_ms": 85.04757, "cand_ms": 99.37774, "n_trials": 12, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 46.15, "compile_s": 1.91, "r_error": "", "src_len": 1035, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 1025, "ts": 1791533663.0256433, "chk": "ok", "code_len": 1035}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6662, "tout": 213, "ts": 1791533679.1121292, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6899, "tout": 516, "ts": 1791533683.9547465, "chk": null, "code_len": 1035}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791533649.2893357}
{"ep": "h2-00048", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0643, "speedup": 1.0643, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.4, "calls": 3, "tokens_in": 23326, "tokens_out": 2737, "ctx_chars": 10527, "t_end": 1791533730.7150457, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0643, "lo95": 1.0609, "mean": 1.0643, "ev_only": 1.0642, "ref_ms": 19.80832, "cand_ms": 18.61197, "n_trials": 78, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0085, "r_wall_s": 41.28, "compile_s": 2.09, "r_error": "", "src_len": 2327, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1511, "ts": 1791533665.4474325, "chk": "ok", "code_len": 2382}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8284, "tout": 73, "ts": 1791533679.6052473, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8381, "tout": 1153, "ts": 1791533689.3909054, "chk": null, "code_len": 2327}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791533649.287733}
{"ep": "h2-00044", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0183, "speedup": 1.0183, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.9, "calls": 3, "tokens_in": 24881, "tokens_out": 2288, "ctx_chars": 12352, "t_end": 1791533731.1739185, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0183, "lo95": 1.0177, "mean": 1.0187, "ev_only": 1.0185, "ref_ms": 64.60384, "cand_ms": 63.44476, "n_trials": 11, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 45.68, "compile_s": 2.0, "r_error": "", "src_len": 1886, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1318, "ts": 1791533663.8539152, "chk": "ok", "code_len": 2000}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8758, "tout": 13, "ts": 1791533677.3991945, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8795, "tout": 957, "ts": 1791533685.4738793, "chk": null, "code_len": 1886}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791533649.2859263}
{"ep": "h2-00037", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7159, "speedup": 0.7159, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 82.4, "calls": 2, "tokens_in": 16327, "tokens_out": 3810, "ctx_chars": 12352, "t_end": 1791533731.724488, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.7159, "lo95": 0.7148, "mean": 0.7166, "ev_only": 0.7154, "ref_ms": 64.34759, "cand_ms": 89.88883, "n_trials": 16, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0009, "r_wall_s": 35.87, "compile_s": 1.92, "r_error": "", "src_len": 2033, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 2731, "ts": 1791533675.278063, "chk": "ok", "code_len": 2115}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8997, "tout": 1079, "ts": 1791533695.7444265, "chk": null, "code_len": 2033}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791533649.2821634}
{"ep": "h2-00058", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0031, "speedup": 1.0031, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 84.3, "calls": 3, "tokens_in": 23960, "tokens_out": 2692, "ctx_chars": 11823, "t_end": 1791533733.6331742, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0031, "lo95": 1.0017, "mean": 1.003, "ev_only": 1.0027, "ref_ms": 27.48936, "cand_ms": 27.4056, "n_trials": 26, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0124, "r_wall_s": 34.27, "compile_s": 2.4, "r_error": "", "src_len": 2347, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1439, "ts": 1791533664.176023, "chk": "ok", "code_len": 2475}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8472, "tout": 71, "ts": 1791533689.6416168, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8567, "tout": 1182, "ts": 1791533699.2592585, "chk": null, "code_len": 2347}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6921, "started": 1791533649.2912676}
{"ep": "h2-00039", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9378, "speedup": 0.9378, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.8, "calls": 2, "tokens_in": 17201, "tokens_out": 4844, "ctx_chars": 11065, "t_end": 1791533740.1108382, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9378, "lo95": 0.9348, "mean": 0.9341, "ev_only": 0.9387, "ref_ms": 1.82346, "cand_ms": 1.94439, "n_trials": 228, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3413, "r_wall_s": 28.87, "compile_s": 2.01, "r_error": "", "src_len": 2047, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 3835, "ts": 1791533688.1995146, "chk": "ok", "code_len": 2047}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10574, "tout": 1009, "ts": 1791533710.7490392, "chk": null, "code_len": 2047}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791533649.283202}
{"ep": "h2-00000", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5494, "speedup": 1.5494, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.9, "calls": 2, "tokens_in": 5188, "tokens_out": 4301, "ctx_chars": 0, "t_end": 1791533740.1600866, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.5494, "lo95": 1.5483, "mean": 1.5479, "ev_only": 1.5472, "ref_ms": 7.99773, "cand_ms": 5.16175, "n_trials": 210, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 25.41, "compile_s": 2.27, "r_error": "", "src_len": 2659, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 2767, "ts": 1791533673.7119293, "chk": "ok", "code_len": 2659}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3587, "tout": 1534, "ts": 1791533714.4469, "chk": null, "code_len": 2659}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791533649.844874}
{"ep": "h2-00059", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.3452, "speedup": 2.3452, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 94.2, "calls": 2, "tokens_in": 13717, "tokens_out": 4437, "ctx_chars": 8529, "t_end": 1791533743.4425619, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.3452, "lo95": 2.3429, "mean": 2.3426, "ev_only": 2.3543, "ref_ms": 8.52004, "cand_ms": 3.63298, "n_trials": 156, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4203, "r_wall_s": 33.79, "compile_s": 2.08, "r_error": "", "src_len": 3516, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2615, "ts": 1791533675.1876233, "chk": "ok", "code_len": 3516}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8222, "tout": 1822, "ts": 1791533709.3794925, "chk": null, "code_len": 3516}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791533649.2918437}
{"ep": "h2-00061", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.016, "speedup": 1.016, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 94.5, "calls": 3, "tokens_in": 25227, "tokens_out": 2507, "ctx_chars": 12352, "t_end": 1791533743.772406, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.016, "lo95": 1.0152, "mean": 1.0473, "ev_only": 1.0164, "ref_ms": 64.62331, "cand_ms": 63.60281, "n_trials": 19, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 50.1, "compile_s": 2.01, "r_error": "", "src_len": 1952, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1491, "ts": 1791533665.3255963, "chk": "ok", "code_len": 1952}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8931, "tout": 13, "ts": 1791533685.24241, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8968, "tout": 1003, "ts": 1791533693.6275632, "chk": null, "code_len": 1952}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791533649.292477}
{"ep": "h2-00057", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9756, "speedup": 0.9756, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 95.6, "calls": 1, "tokens_in": 7032, "tokens_out": 1156, "ctx_chars": 11539, "t_end": 1791533744.8943644, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9756, "lo95": 0.9745, "mean": 0.9751, "ev_only": 0.9759, "ref_ms": 31.38619, "cand_ms": 32.17173, "n_trials": 42, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0209, "r_wall_s": 81.52, "compile_s": 2.02, "r_error": "", "src_len": 1885, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1156, "ts": 1791533662.523989, "chk": null, "code_len": 1885}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791533649.2917361}
{"ep": "h2-00017", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.754, "speedup": 0.754, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 96.3, "calls": 2, "tokens_in": 12111, "tokens_out": 1550, "ctx_chars": 8224, "t_end": 1791533745.5235782, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.754, "lo95": 0.7302, "mean": 0.9145, "ev_only": 0.7536, "ref_ms": 50.44646, "cand_ms": 66.90628, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0004, "r_wall_s": 67.53, "compile_s": 2.99, "r_error": "", "src_len": 1061, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 986, "ts": 1791533661.8240256, "chk": "ok", "code_len": 1061}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6603, "tout": 564, "ts": 1791533677.9452152, "chk": null, "code_len": 1061}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5508, "started": 1791533649.2748852}
{"ep": "h2-00010", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0626, "speedup": 1.0626, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.0, "calls": 1, "tokens_in": 4605, "tokens_out": 1161, "ctx_chars": 6513, "t_end": 1791533746.267331, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0626, "lo95": 1.0612, "mean": 1.0616, "ev_only": 1.063, "ref_ms": 15.48385, "cand_ms": 14.57153, "n_trials": 88, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1327, "r_wall_s": 85.0, "compile_s": 2.01, "r_error": "", "src_len": 2454, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 4605, "tout": 1161, "ts": 1791533660.4417794, "chk": null, "code_len": 2454}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 4605, "started": 1791533649.2704706}
{"ep": "h2-00025", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8849, "speedup": 0.8849, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 99.2, "calls": 2, "tokens_in": 4781, "tokens_out": 2610, "ctx_chars": 0, "t_end": 1791533748.5028522, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.8849, "lo95": 0.7735, "mean": 0.9008, "ev_only": 0.982, "ref_ms": 0.17623, "cand_ms": 0.19916, "n_trials": 210, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 57.97, "compile_s": 11.9, "r_error": "", "src_len": 2339, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 1374, "ts": 1791533661.8267777, "chk": "ok", "code_len": 2339}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3132, "tout": 1236, "ts": 1791533690.4000986, "chk": null, "code_len": 2339}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791533649.278272}
{"ep": "h2-00009", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0152, "speedup": 1.0152, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 102.4, "calls": 3, "tokens_in": 24728, "tokens_out": 2188, "ctx_chars": 12117, "t_end": 1791533751.6785705, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0152, "lo95": 1.0131, "mean": 1.015, "ev_only": 1.0152, "ref_ms": 42.70454, "cand_ms": 42.06629, "n_trials": 31, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 60.71, "compile_s": 1.97, "r_error": "", "src_len": 2260, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1116, "ts": 1791533660.5843143, "chk": "ok", "code_len": 2260}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8625, "tout": 57, "ts": 1791533681.8214827, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8706, "tout": 1015, "ts": 1791533690.333322, "chk": null, "code_len": 2260}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791533649.2700288}
{"ep": "h2-00029", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0054, "speedup": 1.0054, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 103.9, "calls": 3, "tokens_in": 20113, "tokens_out": 2774, "ctx_chars": 9266, "t_end": 1791533753.1778765, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0054, "lo95": 1.0051, "mean": 1.0049, "ev_only": 1.005, "ref_ms": 30.42998, "cand_ms": 30.26667, "n_trials": 59, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0421, "r_wall_s": 46.83, "compile_s": 2.23, "r_error": "", "src_len": 2417, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1472, "ts": 1791533664.6813707, "chk": "ok", "code_len": 2417}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7198, "tout": 79, "ts": 1791533695.5882344, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7301, "tout": 1223, "ts": 1791533705.7646875, "chk": null, "code_len": 2417}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791533649.2793083}
{"ep": "h2-00043", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0093, "speedup": 1.0093, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 104.7, "calls": 3, "tokens_in": 25332, "tokens_out": 2590, "ctx_chars": 12352, "t_end": 1791533754.0171568, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0093, "lo95": 1.0078, "mean": 1.0087, "ev_only": 1.0096, "ref_ms": 64.27465, "cand_ms": 63.68273, "n_trials": 13, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0025, "r_wall_s": 50.74, "compile_s": 2.69, "r_error": "", "src_len": 1982, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1530, "ts": 1791533665.909805, "chk": "ok", "code_len": 2131}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8970, "tout": 40, "ts": 1791533694.2634296, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9034, "tout": 1020, "ts": 1791533703.0283582, "chk": null, "code_len": 1982}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791533649.2862167}
{"ep": "h2-00038", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.325, "speedup": 1.325, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 106.5, "calls": 1, "tokens_in": 5998, "tokens_out": 735, "ctx_chars": 9970, "t_end": 1791533755.8085637, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.325, "lo95": 1.3236, "mean": 1.3257, "ev_only": 1.3301, "ref_ms": 7.86798, "cand_ms": 5.93828, "n_trials": 184, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0902, "r_wall_s": 96.0, "compile_s": 1.99, "r_error": "", "src_len": 1433, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 735, "ts": 1791533659.173716, "chk": null, "code_len": 1433}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791533649.2830698}
{"ep": "h2-00019", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.01, "speedup": 1.01, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 107.1, "calls": 2, "tokens_in": 16203, "tokens_out": 2577, "ctx_chars": 12117, "t_end": 1791533756.3399673, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.01, "lo95": 1.0087, "mean": 1.01, "ev_only": 1.0096, "ref_ms": 42.66755, "cand_ms": 42.24501, "n_trials": 38, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 66.53, "compile_s": 2.0, "r_error": "", "src_len": 2064, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1561, "ts": 1791533663.6413333, "chk": "ok", "code_len": 2064}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8804, "tout": 1016, "ts": 1791533689.036278, "chk": null, "code_len": 2064}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791533649.275489}
{"ep": "h2-00011", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0514, "speedup": 1.0514, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 108.8, "calls": 3, "tokens_in": 18684, "tokens_out": 3801, "ctx_chars": 6513, "t_end": 1791533758.0351768, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0514, "lo95": 1.0483, "mean": 1.0541, "ev_only": 1.0526, "ref_ms": 15.79707, "cand_ms": 15.02523, "n_trials": 83, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.144, "r_wall_s": 48.5, "compile_s": 2.65, "r_error": "", "src_len": 3782, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1891, "ts": 1791533665.7900488, "chk": "ok", "code_len": 3782}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6865, "tout": 68, "ts": 1791533694.9577398, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6957, "tout": 1842, "ts": 1791533709.3813062, "chk": null, "code_len": 3782}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791533649.2714355}
{"ep": "h2-00002", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0049, "speedup": 1.0049, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 109.4, "calls": 1, "tokens_in": 1623, "tokens_out": 2097, "ctx_chars": 0, "t_end": 1791533758.7138653, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0049, "lo95": 1.0046, "mean": 1.0047, "ev_only": 1.005, "ref_ms": 30.50454, "cand_ms": 30.35692, "n_trials": 45, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0432, "r_wall_s": 89.34, "compile_s": 47.32, "r_error": "", "src_len": 4427, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1623, "tout": 2097, "ts": 1791533668.856897, "chk": null, "code_len": 4427}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1623, "started": 1791533649.2662203}
{"ep": "h2-00014", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9713, "speedup": 0.9713, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.6, "calls": 2, "tokens_in": 12255, "tokens_out": 3112, "ctx_chars": 9033, "t_end": 1791533760.8258991, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9713, "lo95": 0.971, "mean": 0.9711, "ev_only": 0.973, "ref_ms": 15.52485, "cand_ms": 15.98325, "n_trials": 79, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.014, "r_wall_s": 53.82, "compile_s": 2.11, "r_error": "", "src_len": 1455, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 2078, "ts": 1791533670.0603368, "chk": "ok", "code_len": 1455}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6765, "tout": 1034, "ts": 1791533706.6575987, "chk": null, "code_len": 1455}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791533649.2738817}
{"ep": "h2-00050", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9979, "speedup": 0.9979, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.8, "calls": 3, "tokens_in": 23097, "tokens_out": 3438, "ctx_chars": 10823, "t_end": 1791533761.1211636, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9979, "lo95": 0.996, "mean": 0.9968, "ev_only": 0.9967, "ref_ms": 30.17001, "cand_ms": 30.23255, "n_trials": 50, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0385, "r_wall_s": 62.57, "compile_s": 2.04, "r_error": "", "src_len": 3170, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1860, "ts": 1791533668.0461526, "chk": "ok", "code_len": 3373}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8325, "tout": 70, "ts": 1791533684.1553066, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8419, "tout": 1508, "ts": 1791533698.0910246, "chk": null, "code_len": 3170}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791533649.2884767}
{"ep": "h2-00023", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9793, "speedup": 0.9793, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 113.0, "calls": 2, "tokens_in": 12519, "tokens_out": 3656, "ctx_chars": 9033, "t_end": 1791533762.316682, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.9793, "lo95": 0.9779, "mean": 1.0071, "ev_only": 0.981, "ref_ms": 15.50906, "cand_ms": 15.83645, "n_trials": 73, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.015, "r_wall_s": 62.55, "compile_s": 2.0, "r_error": "", "src_len": 1235, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 2533, "ts": 1791533674.8252478, "chk": "ok", "code_len": 1764}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7029, "tout": 1123, "ts": 1791533699.4509578, "chk": null, "code_len": 1235}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791533649.2762918}
{"ep": "h2-00068", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7208, "speedup": 0.7208, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 74.5, "calls": 1, "tokens_in": 6622, "tokens_out": 505, "ctx_chars": 10850, "t_end": 1791533778.402674, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.7208, "lo95": 0.699, "mean": 0.711, "ev_only": 0.7371, "ref_ms": 0.15503, "cand_ms": 0.21507, "n_trials": 210, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 68.92, "compile_s": 2.0, "r_error": "", "src_len": 891, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6622, "tout": 505, "ts": 1791533709.3576987, "chk": null, "code_len": 891}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6622, "started": 1791533703.9079251}
{"ep": "h2-00070", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 71.1, "calls": 1, "tokens_in": 2181, "tokens_out": 1022, "ctx_chars": 0, "t_end": 1791533781.1865263, "error": "RuntimeError: Error building extension 'pt_passthrough': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=pt_passthrough -DTORCH_API_INCLUDE_EXTENSION_H -I/us", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 59.25, "compile_s": 59.25, "r_error": "RuntimeError: Error building extension 'pt_passthrough': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=pt_passthrough -DTORCH_API_INCLUDE_EXTENSION_H -I/us", "src_len": 1787, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2181, "tout": 1022, "ts": 1791533721.8697422, "chk": null, "code_len": 1787}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2181, "started": 1791533710.6600232}
{"ep": "h2-00063", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.4591, "speedup": 1.4591, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 139.5, "calls": 2, "tokens_in": 11216, "tokens_out": 5074, "ctx_chars": 6513, "t_end": 1791533788.7697291, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.4591, "lo95": 1.4583, "mean": 1.4575, "ev_only": 1.4603, "ref_ms": 15.50178, "cand_ms": 10.62413, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1675, "r_wall_s": 76.05, "compile_s": 2.14, "r_error": "", "src_len": 3357, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3556, "ts": 1791533682.6013086, "chk": "ok", "code_len": 3357}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6609, "tout": 1518, "ts": 1791533711.3693218, "chk": null, "code_len": 3357}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791533649.2993574}
{"ep": "h2-00052", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.993, "speedup": 0.993, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 140.2, "calls": 3, "tokens_in": 26546, "tokens_out": 5371, "ctx_chars": 10823, "t_end": 1791533789.4504995, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.993, "lo95": 0.9913, "mean": 0.9931, "ev_only": 0.994, "ref_ms": 30.14962, "cand_ms": 30.3607, "n_trials": 46, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0331, "r_wall_s": 79.99, "compile_s": 2.03, "r_error": "", "src_len": 3512, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 3588, "ts": 1791533680.9534407, "chk": "ok", "code_len": 3512}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10053, "tout": 63, "ts": 1791533694.8519192, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10140, "tout": 1720, "ts": 1791533708.7585318, "chk": null, "code_len": 3512}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791533649.289909}
{"ep": "h2-00066", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7122, "speedup": 0.7122, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 105.7, "calls": 2, "tokens_in": 16364, "tokens_out": 3856, "ctx_chars": 12352, "t_end": 1791533793.148018, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.7122, "lo95": 0.7105, "mean": 0.7119, "ev_only": 0.7122, "ref_ms": 64.12535, "cand_ms": 90.03768, "n_trials": 17, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.002, "r_wall_s": 13.41, "compile_s": 1.97, "r_error": "", "src_len": 2104, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 2707, "ts": 1791533712.202178, "chk": "ok", "code_len": 2104}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9034, "tout": 1149, "ts": 1791533779.6919744, "chk": null, "code_len": 2104}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791533687.4799795}
{"ep": "h2-00026", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0653, "speedup": 1.0653, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 145.7, "calls": 2, "tokens_in": 10510, "tokens_out": 2310, "ctx_chars": 6513, "t_end": 1791533795.010232, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0653, "lo95": 1.0651, "mean": 1.0646, "ev_only": 1.0647, "ref_ms": 15.48248, "cand_ms": 14.53333, "n_trials": 97, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1323, "r_wall_s": 102.02, "compile_s": 2.37, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1188, "ts": 1791533660.8327847, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5905, "tout": 1122, "ts": 1791533692.2913747, "chk": null, "code_len": 2422}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791533649.2776418}
{"ep": "h2-00055", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9926, "speedup": 0.9926, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 153.7, "calls": 3, "tokens_in": 24790, "tokens_out": 4031, "ctx_chars": 10167, "t_end": 1791533802.9974284, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9926, "lo95": 0.9903, "mean": 0.9898, "ev_only": 0.9983, "ref_ms": 8.54362, "cand_ms": 8.60688, "n_trials": 194, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0573, "r_wall_s": 86.24, "compile_s": 2.4, "r_error": "", "src_len": 1840, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3083, "ts": 1791533681.075079, "chk": "ok", "code_len": 1840}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9299, "tout": 64, "ts": 1791533708.8919163, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9387, "tout": 884, "ts": 1791533716.409915, "chk": null, "code_len": 1840}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791533649.2904763}
{"ep": "h2-00041", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0258, "speedup": 1.0258, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 157.8, "calls": 3, "tokens_in": 15531, "tokens_out": 6423, "ctx_chars": 5026, "t_end": 1791533807.075003, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0258, "lo95": 1.0255, "mean": 1.0254, "ev_only": 1.0264, "ref_ms": 8.18186, "cand_ms": 7.97589, "n_trials": 136, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0072, "r_wall_s": 17.28, "compile_s": 2.01, "r_error": "", "src_len": 1426, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1742, "ts": 1791533666.758341, "chk": "incorrect", "code_len": 1337}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5232, "tout": 3817, "ts": 1791533731.8509443, "chk": "ok", "code_len": 1426}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6304, "tout": 864, "ts": 1791533789.7275255, "chk": null, "code_len": 1426}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", null], "first_tin": 3995, "started": 1791533649.2861032}
{"ep": "h2-00031", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.6675, "speedup": 3.6675, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 163.4, "calls": 2, "tokens_in": 6074, "tokens_out": 9514, "ctx_chars": 0, "t_end": 1791533812.6669052, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 3.6675, "lo95": 3.6661, "mean": 3.6655, "ev_only": 3.6708, "ref_ms": 64.34386, "cand_ms": 17.5443, "n_trials": 26, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0079, "r_wall_s": 12.24, "compile_s": 1.82, "r_error": "", "src_len": 3251, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 7764, "ts": 1791533725.8790867, "chk": "ok", "code_len": 3251}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4328, "tout": 1750, "ts": 1791533800.3155618, "chk": null, "code_len": 3251}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791533649.2800713}
{"ep": "h2-00006", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0067, "speedup": 1.0067, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 166.9, "calls": 2, "tokens_in": 5699, "tokens_out": 5876, "ctx_chars": 0, "t_end": 1791533816.1408408, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0067, "lo95": 1.0049, "mean": 1.0064, "ev_only": 1.0081, "ref_ms": 30.1169, "cand_ms": 29.9155, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0204, "r_wall_s": 98.84, "compile_s": 2.79, "r_error": "", "src_len": 4001, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 3880, "ts": 1791533683.6057248, "chk": "ok", "code_len": 4072}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4101, "tout": 1996, "ts": 1791533716.6291323, "chk": null, "code_len": 4001}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791533649.269165}
{"ep": "h2-00032", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.7808, "speedup": 2.7808, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 169.4, "calls": 2, "tokens_in": 13906, "tokens_out": 4387, "ctx_chars": 9970, "t_end": 1791533818.6598034, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.7808, "lo95": 2.7781, "mean": 2.7764, "ev_only": 2.7936, "ref_ms": 7.84245, "cand_ms": 2.82023, "n_trials": 181, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 101.02, "compile_s": 2.85, "r_error": "", "src_len": 2640, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 3073, "ts": 1791533677.3916428, "chk": "ok", "code_len": 2640}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7906, "tout": 1314, "ts": 1791533716.7555034, "chk": null, "code_len": 2640}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791533649.2804623}
{"ep": "h2-00035", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.6064, "speedup": 0.6064, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 169.4, "calls": 2, "tokens_in": 15280, "tokens_out": 11050, "ctx_chars": 6513, "t_end": 1791533818.6958876, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.6064, "lo95": 0.6058, "mean": 0.6071, "ev_only": 0.6059, "ref_ms": 7.13591, "cand_ms": 11.76748, "n_trials": 127, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3802, "r_wall_s": 16.25, "compile_s": 1.89, "r_error": "", "src_len": 7560, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 7515, "ts": 1791533714.0259528, "chk": "ok", "code_len": 7594}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9633, "tout": 3535, "ts": 1791533802.4185514, "chk": null, "code_len": 7560}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791533649.281854}
{"ep": "h2-00008", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7915, "speedup": 0.7915, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 172.9, "calls": 3, "tokens_in": 50140, "tokens_out": 14983, "ctx_chars": 2425, "t_end": 1791533822.2172344, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7915, "lo95": 0.7891, "mean": 0.8117, "ev_only": 0.7923, "ref_ms": 18.79661, "cand_ms": 23.74955, "n_trials": 48, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0299, "r_wall_s": 18.12, "compile_s": 1.84, "r_error": "", "src_len": 17465, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7601, "ts": 1791533711.5850685, "chk": "ok", "code_len": 17465}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19261, "tout": 46, "ts": 1791533747.0442054, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19331, "tout": 7336, "ts": 1791533804.0626333, "chk": null, "code_len": 17465}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791533649.2700546}
{"ep": "h2-00060", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0176, "speedup": 1.0176, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 173.7, "calls": 2, "tokens_in": 15387, "tokens_out": 3233, "ctx_chars": 11823, "t_end": 1791533822.9989917, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0176, "lo95": 1.0164, "mean": 1.0183, "ev_only": 1.0173, "ref_ms": 27.5325, "cand_ms": 27.05626, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.015, "r_wall_s": 77.67, "compile_s": 1.94, "r_error": "", "src_len": 2024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 2140, "ts": 1791533669.322927, "chk": "ok", "code_len": 2024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8464, "tout": 1093, "ts": 1791533745.2866735, "chk": null, "code_len": 2024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791533649.2926145}
{"ep": "h2-00077", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0081, "speedup": 1.0081, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.7, "calls": 1, "tokens_in": 6367, "tokens_out": 545, "ctx_chars": 10847, "t_end": 1791533827.7004654, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0081, "lo95": 1.0074, "mean": 1.0342, "ev_only": 1.0068, "ref_ms": 7.96362, "cand_ms": 7.89949, "n_trials": 138, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 74.3, "compile_s": 2.82, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 545, "ts": 1791533753.1030674, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791533729.9674354}
{"ep": "h2-00099", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0625, "speedup": 1.0625, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 68.3, "calls": 2, "tokens_in": 4925, "tokens_out": 4057, "ctx_chars": 0, "t_end": 1791533829.1418555, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0625, "lo95": 1.0621, "mean": 1.0621, "ev_only": 1.0631, "ref_ms": 8.4912, "cand_ms": 7.9917, "n_trials": 136, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0052, "r_wall_s": 17.75, "compile_s": 2.03, "r_error": "", "src_len": 2170, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 2822, "ts": 1791533786.2686498, "chk": "ok", "code_len": 2170}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3290, "tout": 1235, "ts": 1791533811.3186412, "chk": null, "code_len": 2170}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791533760.8261466}
{"ep": "h2-00065", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0069, "speedup": 1.0069, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 150.8, "calls": 2, "tokens_in": 4118, "tokens_out": 2263, "ctx_chars": 0, "t_end": 1791533832.154856, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0069, "lo95": 1.0059, "mean": 1.008, "ev_only": 1.0058, "ref_ms": 8.01812, "cand_ms": 7.96299, "n_trials": 142, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 80.19, "compile_s": 2.86, "r_error": "", "src_len": 1027, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1615, "ts": 1791533695.7978497, "chk": "ok", "code_len": 1027}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2615, "tout": 648, "ts": 1791533751.6727138, "chk": null, "code_len": 1027}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791533681.351586}
{"ep": "h2-00095", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0612, "speedup": 1.0612, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 76.8, "calls": 2, "tokens_in": 14982, "tokens_out": 3562, "ctx_chars": 10167, "t_end": 1791533832.582637, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0612, "lo95": 1.0597, "mean": 1.057, "ev_only": 1.062, "ref_ms": 8.53501, "cand_ms": 8.04316, "n_trials": 145, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 30.01, "compile_s": 1.98, "r_error": "", "src_len": 1781, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 2662, "ts": 1791533783.5403523, "chk": "ok", "code_len": 1781}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8878, "tout": 900, "ts": 1791533802.498028, "chk": null, "code_len": 1781}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6104, "started": 1791533755.8088586}
{"ep": "h2-00103", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 56.0, "calls": 1, "tokens_in": 1723, "tokens_out": 1115, "ctx_chars": 0, "t_end": 1791533837.2187164, "error": "RuntimeError: Error building extension 'fuse_conv3d_act': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fuse_conv3d_act -DTORCH_API_INCLUDE_EXTENSION_H -I/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 45.67, "compile_s": 45.66, "r_error": "RuntimeError: Error building extension 'fuse_conv3d_act': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fuse_conv3d_act -DTORCH_API_INCLUDE_EXTENSION_H -I/", "src_len": 2491, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1723, "tout": 1115, "ts": 1791533791.52494, "chk": null, "code_len": 2491}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1723, "started": 1791533781.1868005}
{"ep": "h2-00107", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0277, "speedup": 1.0277, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 47.7, "calls": 2, "tokens_in": 4760, "tokens_out": 2466, "ctx_chars": 0, "t_end": 1791533842.6872501, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0277, "lo95": 1.0273, "mean": 1.0275, "ev_only": 1.0281, "ref_ms": 8.14473, "cand_ms": 7.92486, "n_trials": 148, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0072, "r_wall_s": 16.66, "compile_s": 2.02, "r_error": "", "src_len": 1409, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 1667, "ts": 1791533811.5623772, "chk": "ok", "code_len": 1409}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3008, "tout": 799, "ts": 1791533825.8790848, "chk": null, "code_len": 1409}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1752, "started": 1791533795.0104961}
{"ep": "h2-00089", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 5.509, "speedup": 5.509, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.4, "calls": 2, "tokens_in": 5935, "tokens_out": 8777, "ctx_chars": 0, "t_end": 1791533845.915913, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 5.509, "lo95": 5.5029, "mean": 5.4992, "ev_only": 5.5145, "ref_ms": 64.38748, "cand_ms": 11.68768, "n_trials": 24, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0356, "r_wall_s": 12.37, "compile_s": 1.95, "r_error": "", "src_len": 3359, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 6944, "ts": 1791533810.582562, "chk": "ok", "code_len": 3394}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4189, "tout": 1833, "ts": 1791533833.5258207, "chk": null, "code_len": 3359}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791533745.523831}
{"ep": "h2-00100", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.5412, "speedup": 0.5412, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 85.9, "calls": 3, "tokens_in": 21192, "tokens_out": 1010, "ctx_chars": 10850, "t_end": 1791533846.9826348, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.5412, "lo95": 0.5353, "mean": 0.594, "ev_only": 0.5912, "ref_ms": 0.15402, "cand_ms": 0.28459, "n_trials": 200, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 28.69, "compile_s": 1.84, "r_error": "", "src_len": 818, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 510, "ts": 1791533766.9226902, "chk": "ok", "code_len": 818}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7241, "tout": 64, "ts": 1791533813.7953029, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7329, "tout": 436, "ts": 1791533818.0840597, "chk": null, "code_len": 818}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6622, "started": 1791533761.1214097}
{"ep": "h2-00098", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.4207, "speedup": 2.4207, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.8, "calls": 2, "tokens_in": 12928, "tokens_out": 3883, "ctx_chars": 8529, "t_end": 1791533849.5308928, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 2.4207, "lo95": 2.4201, "mean": 2.4214, "ev_only": 2.4286, "ref_ms": 8.49628, "cand_ms": 3.50988, "n_trials": 193, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4023, "r_wall_s": 18.53, "compile_s": 1.84, "r_error": "", "src_len": 2822, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2317, "ts": 1791533778.2780237, "chk": "ok", "code_len": 2822}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7431, "tout": 1566, "ts": 1791533830.965498, "chk": null, "code_len": 2822}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791533758.7141948}
{"ep": "h2-00012", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0196, "speedup": 1.0196, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 204.9, "calls": 3, "tokens_in": 13281, "tokens_out": 8961, "ctx_chars": 0, "t_end": 1791533854.183165, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0196, "lo95": 1.0187, "mean": 1.0213, "ev_only": 1.021, "ref_ms": 30.47915, "cand_ms": 29.89415, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0169, "r_wall_s": 114.95, "compile_s": 2.4, "r_error": "", "src_len": 4413, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 4167, "ts": 1791533684.20107, "chk": "forbidden", "code_len": 4388}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4594, "tout": 2536, "ts": 1791533702.0239975, "chk": "ok", "code_len": 4413}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7062, "tout": 2258, "ts": 1791533738.6262014, "chk": null, "code_len": 4413}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", null], "first_tin": 1625, "started": 1791533649.2715802}
{"ep": "h2-00020", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3666, "speedup": 1.3666, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 208.5, "calls": 2, "tokens_in": 16394, "tokens_out": 7138, "ctx_chars": 11539, "t_end": 1791533857.7544677, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.3666, "lo95": 1.3656, "mean": 1.3673, "ev_only": 1.3678, "ref_ms": 31.36885, "cand_ms": 22.95376, "n_trials": 49, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9604, "r_wall_s": 98.0, "compile_s": 2.41, "r_error": "", "src_len": 3626, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 5188, "ts": 1791533695.9237363, "chk": "ok", "code_len": 3626}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9360, "tout": 1950, "ts": 1791533758.75671, "chk": null, "code_len": 3626}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791533649.2764084}
{"ep": "h2-00102", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0079, "speedup": 1.0079, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 83.5, "calls": 2, "tokens_in": 18245, "tokens_out": 4539, "ctx_chars": 12352, "t_end": 1791533861.9033878, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0079, "lo95": 1.0075, "mean": 1.0087, "ev_only": 1.0078, "ref_ms": 64.33206, "cand_ms": 63.82856, "n_trials": 19, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0032, "r_wall_s": 32.59, "compile_s": 2.16, "r_error": "", "src_len": 2025, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 3477, "ts": 1791533809.5000205, "chk": "ok", "code_len": 2025}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10917, "tout": 1062, "ts": 1791533829.2493212, "chk": null, "code_len": 2025}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7328, "started": 1791533778.9416595}
{"ep": "h2-00105", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0528, "speedup": 1.0528, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 75.6, "calls": 2, "tokens_in": 11449, "tokens_out": 3175, "ctx_chars": 6513, "t_end": 1791533865.0790184, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0528, "lo95": 1.0517, "mean": 1.0707, "ev_only": 1.0529, "ref_ms": 15.74662, "cand_ms": 14.95656, "n_trials": 105, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1724, "r_wall_s": 34.27, "compile_s": 2.12, "r_error": "", "src_len": 3280, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1613, "ts": 1791533804.1907954, "chk": "ok", "code_len": 3280}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6587, "tout": 1562, "ts": 1791533830.6235285, "chk": null, "code_len": 3280}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4862, "started": 1791533789.4507468}
{"ep": "h2-00024", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0695, "speedup": 1.0695, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 217.8, "calls": 2, "tokens_in": 28562, "tokens_out": 17943, "ctx_chars": 0, "t_end": 1791533867.0345907, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0695, "lo95": 1.0674, "mean": 1.07, "ev_only": 1.071, "ref_ms": 18.97954, "cand_ms": 17.74685, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3765, "r_wall_s": 23.29, "compile_s": 1.98, "r_error": "", "src_len": 15042, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 11141, "ts": 1791533737.471595, "chk": "ok", "code_len": 15071}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 17928, "tout": 6802, "ts": 1791533843.7115743, "chk": null, "code_len": 15042}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791533649.277822}
{"ep": "h2-00064", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.034, "speedup": 1.034, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 189.5, "calls": 4, "tokens_in": 19950, "tokens_out": 5004, "ctx_chars": 0, "t_end": 1791533868.0951405, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.034, "lo95": 1.0337, "mean": 1.0693, "ev_only": 1.0344, "ref_ms": 8.29078, "cand_ms": 8.01852, "n_trials": 151, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0145, "r_wall_s": 47.28, "compile_s": 11.08, "r_error": "", "src_len": 3547, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1764, "tout": 1909, "ts": 1791533697.448191, "chk": "compile_error", "code_len": 4028}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4943, "tout": 1548, "ts": 1791533762.2305653, "chk": "ok", "code_len": 3547}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6603, "tout": 13, "ts": 1791533808.0912547, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6640, "tout": 1534, "ts": 1791533820.6527827, "chk": null, "code_len": 3547}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 1764, "started": 1791533678.6241403}
{"ep": "h2-00113", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0012, "speedup": 1.0012, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 53.9, "calls": 2, "tokens_in": 13463, "tokens_out": 1136, "ctx_chars": 10847, "t_end": 1791533872.591003, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0012, "lo95": 1.0007, "mean": 1.0012, "ev_only": 1.0018, "ref_ms": 7.94758, "cand_ms": 7.93816, "n_trials": 148, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 30.33, "compile_s": 1.88, "r_error": "", "src_len": 823, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 616, "ts": 1791533824.5194488, "chk": "ok", "code_len": 823}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7094, "tout": 520, "ts": 1791533842.1910748, "chk": null, "code_len": 823}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791533818.6960862}
{"ep": "h2-00101", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0079, "speedup": 1.0079, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 115.0, "calls": 3, "tokens_in": 19853, "tokens_out": 2128, "ctx_chars": 9274, "t_end": 1791533877.317239, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0079, "lo95": 1.0059, "mean": 1.0129, "ev_only": 1.0075, "ref_ms": 8.3594, "cand_ms": 8.29347, "n_trials": 135, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0439, "r_wall_s": 29.01, "compile_s": 1.96, "r_error": "", "src_len": 1954, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1087, "ts": 1791533772.5696986, "chk": "ok", "code_len": 2012}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6989, "tout": 61, "ts": 1791533840.0500696, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7074, "tout": 980, "ts": 1791533848.2271237, "chk": null, "code_len": 1954}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791533762.3169558}
{"ep": "h2-00072", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0406, "speedup": 1.0406, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 166.3, "calls": 2, "tokens_in": 13177, "tokens_out": 2611, "ctx_chars": 9274, "t_end": 1791533878.8825011, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0406, "lo95": 1.0399, "mean": 1.0397, "ev_only": 1.0431, "ref_ms": 8.33593, "cand_ms": 8.01105, "n_trials": 162, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0068, "r_wall_s": 105.93, "compile_s": 1.97, "r_error": "", "src_len": 2024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1481, "ts": 1791533724.5476148, "chk": "ok", "code_len": 2024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7385, "tout": 1130, "ts": 1791533772.647507, "chk": null, "code_len": 2024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791533712.5361474}
{"ep": "h2-00054", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9329, "speedup": 0.9329, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 229.6, "calls": 2, "tokens_in": 14488, "tokens_out": 8404, "ctx_chars": 6513, "t_end": 1791533878.8919432, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9329, "lo95": 0.9273, "mean": 0.8739, "ev_only": 0.933, "ref_ms": 3.30113, "cand_ms": 3.53874, "n_trials": 154, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0773, "r_wall_s": 107.22, "compile_s": 3.38, "r_error": "", "src_len": 6642, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 5261, "ts": 1791533694.2967224, "chk": "ok", "code_len": 6642}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9040, "tout": 3143, "ts": 1791533771.6245291, "chk": null, "code_len": 6642}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791533649.2907917}
{"ep": "h2-00106", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.017, "speedup": 1.017, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 86.9, "calls": 2, "tokens_in": 15594, "tokens_out": 3719, "ctx_chars": 11823, "t_end": 1791533880.011174, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.017, "lo95": 1.0156, "mean": 1.0194, "ev_only": 1.0186, "ref_ms": 27.7672, "cand_ms": 27.30221, "n_trials": 55, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0144, "r_wall_s": 41.86, "compile_s": 2.08, "r_error": "", "src_len": 2082, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 2573, "ts": 1791533820.040803, "chk": "ok", "code_len": 2082}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8671, "tout": 1146, "ts": 1791533838.1298468, "chk": null, "code_len": 2082}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791533793.14827}
{"ep": "h2-00086", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0181, "speedup": 1.0181, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 150.8, "calls": 2, "tokens_in": 13639, "tokens_out": 5566, "ctx_chars": 9266, "t_end": 1791533894.2102656, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0181, "lo95": 1.016, "mean": 1.0185, "ev_only": 1.0184, "ref_ms": 30.61966, "cand_ms": 30.07637, "n_trials": 36, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0156, "r_wall_s": 58.24, "compile_s": 1.95, "r_error": "", "src_len": 3662, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 3722, "ts": 1791533775.8972828, "chk": "ok", "code_len": 3662}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8023, "tout": 1844, "ts": 1791533835.522458, "chk": null, "code_len": 3662}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5616, "started": 1791533743.442833}
{"ep": "h2-00115", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 72.0, "calls": 1, "tokens_in": 1666, "tokens_out": 767, "ctx_chars": 0, "t_end": 1791533894.9663093, "error": "RuntimeError: Error building extension 'mish_tanh_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mish_tanh_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 64.34, "compile_s": 64.34, "r_error": "RuntimeError: Error building extension 'mish_tanh_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mish_tanh_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/", "src_len": 1523, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1666, "tout": 767, "ts": 1791533830.5831773, "chk": null, "code_len": 1523}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1666, "started": 1791533822.9992323}
{"ep": "h2-00073", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0065, "speedup": 1.0065, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 181.9, "calls": 2, "tokens_in": 4211, "tokens_out": 2139, "ctx_chars": 0, "t_end": 1791533895.6988146, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0065, "lo95": 1.0061, "mean": 1.038, "ev_only": 1.0065, "ref_ms": 7.98717, "cand_ms": 7.93555, "n_trials": 172, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 41.74, "compile_s": 1.95, "r_error": "", "src_len": 879, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1638, "ts": 1791533729.4681106, "chk": "ok", "code_len": 879}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2610, "tout": 501, "ts": 1791533853.8820515, "chk": null, "code_len": 879}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791533713.8203542}
{"ep": "h2-00097", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8867, "speedup": 0.8867, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 138.9, "calls": 2, "tokens_in": 14641, "tokens_out": 10298, "ctx_chars": 6513, "t_end": 1791533896.897302, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8867, "lo95": 0.8774, "mean": 0.9263, "ev_only": 0.8833, "ref_ms": 3.5442, "cand_ms": 3.99704, "n_trials": 206, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0708, "r_wall_s": 42.72, "compile_s": 2.01, "r_error": "", "src_len": 6667, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 7067, "ts": 1791533817.9857807, "chk": "ok", "code_len": 6667}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9193, "tout": 3231, "ts": 1791533854.1524434, "chk": null, "code_len": 6667}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791533758.035417}
{"ep": "h2-00090", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.6346, "speedup": 2.6346, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 152.3, "calls": 2, "tokens_in": 5844, "tokens_out": 6529, "ctx_chars": 0, "t_end": 1791533898.519243, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.6346, "lo95": 2.63, "mean": 2.6336, "ev_only": 2.6304, "ref_ms": 7.82839, "cand_ms": 2.97141, "n_trials": 180, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9972, "r_wall_s": 71.43, "compile_s": 12.2, "r_error": "", "src_len": 3944, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 4520, "ts": 1791533785.9059887, "chk": "ok", "code_len": 3944}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4119, "tout": 2009, "ts": 1791533826.426309, "chk": null, "code_len": 3944}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1725, "started": 1791533746.2675757}
{"ep": "h2-00084", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8551, "speedup": 0.8551, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 166.5, "calls": 1, "tokens_in": 7030, "tokens_out": 1213, "ctx_chars": 11175, "t_end": 1791533906.6439242, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8551, "lo95": 0.8488, "mean": 0.8907, "ev_only": 0.8513, "ref_ms": 14.4647, "cand_ms": 16.91527, "n_trials": 62, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 153.75, "compile_s": 2.47, "r_error": "", "src_len": 2558, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1213, "ts": 1791533752.5733333, "chk": null, "code_len": 2558}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791533740.1111097}
{"ep": "h2-00018", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9987, "speedup": 0.9987, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 257.4, "calls": 4, "tokens_in": 34465, "tokens_out": 3133, "ctx_chars": 11539, "t_end": 1791533906.7203655, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9987, "lo95": 0.9982, "mean": 0.9984, "ev_only": 0.9997, "ref_ms": 31.41472, "cand_ms": 31.45524, "n_trials": 21, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 140.55, "compile_s": 2.06, "r_error": "", "src_len": 1822, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1247, "ts": 1791533664.2654042, "chk": "ok", "code_len": 1927}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8388, "tout": 985, "ts": 1791533703.2207818, "chk": "ok", "code_len": 1822}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9482, "tout": 57, "ts": 1791533758.0232227, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 9563, "tout": 844, "ts": 1791533765.1499443, "chk": null, "code_len": 1822}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 7032, "started": 1791533649.275926}
{"ep": "h2-00003", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.8562, "speedup": 2.8562, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 258.1, "calls": 2, "tokens_in": 6961, "tokens_out": 8538, "ctx_chars": 0, "t_end": 1791533907.354147, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.8562, "lo95": 2.8529, "mean": 2.851, "ev_only": 2.8587, "ref_ms": 42.56659, "cand_ms": 14.9032, "n_trials": 41, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9996, "r_wall_s": 149.45, "compile_s": 2.14, "r_error": "", "src_len": 4351, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 6188, "ts": 1791533704.4486878, "chk": "ok", "code_len": 4440}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4925, "tout": 2350, "ts": 1791533756.5832355, "chk": null, "code_len": 4351}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2036, "started": 1791533649.267133}
{"ep": "h2-00075", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0085, "speedup": 1.0085, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 194.1, "calls": 1, "tokens_in": 6477, "tokens_out": 515, "ctx_chars": 10909, "t_end": 1791533909.9205499, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0085, "lo95": 1.0035, "mean": 1.0115, "ev_only": 1.0058, "ref_ms": 8.13953, "cand_ms": 8.07093, "n_trials": 127, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0019, "r_wall_s": 188.48, "compile_s": 3.27, "r_error": "", "src_len": 966, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 515, "ts": 1791533721.0228088, "chk": null, "code_len": 966}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791533715.812501}
{"ep": "h2-00093", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8136, "speedup": 0.8136, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 156.8, "calls": 1, "tokens_in": 1647, "tokens_out": 445, "ctx_chars": 0, "t_end": 1791533910.0007112, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8136, "lo95": 0.7857, "mean": 0.7246, "ev_only": 0.7527, "ref_ms": 0.17947, "cand_ms": 0.22058, "n_trials": 202, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 151.58, "compile_s": 2.52, "r_error": "", "src_len": 785, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1647, "tout": 445, "ts": 1791533758.271849, "chk": null, "code_len": 785}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1647, "started": 1791533753.1781223}
{"ep": "h2-00033", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.1788, "speedup": 4.1788, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 261.4, "calls": 2, "tokens_in": 8256, "tokens_out": 12231, "ctx_chars": 0, "t_end": 1791533910.6763413, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 4.1788, "lo95": 4.1708, "mean": 4.1885, "ev_only": 4.2054, "ref_ms": 8.56624, "cand_ms": 2.04993, "n_trials": 185, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.2324, "r_wall_s": 132.73, "compile_s": 13.37, "r_error": "", "src_len": 8249, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 8251, "ts": 1791533722.6218457, "chk": "ok", "code_len": 8249}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6436, "tout": 3980, "ts": 1791533777.6887014, "chk": null, "code_len": 8249}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791533649.281598}
{"ep": "h2-00087", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0503, "speedup": 1.0503, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 175.5, "calls": 1, "tokens_in": 4605, "tokens_out": 1154, "ctx_chars": 6513, "t_end": 1791533919.246128, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0503, "lo95": 1.0407, "mean": 1.0485, "ev_only": 1.0518, "ref_ms": 15.59323, "cand_ms": 14.84699, "n_trials": 80, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.133, "r_wall_s": 164.67, "compile_s": 2.72, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 4605, "tout": 1154, "ts": 1791533753.680677, "chk": null, "code_len": 2422}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 4605, "started": 1791533743.7726505}
{"ep": "h2-00074", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.8291, "speedup": 0.8291, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 212.5, "calls": 2, "tokens_in": 5812, "tokens_out": 5692, "ctx_chars": 0, "t_end": 1791533927.5179422, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8291, "lo95": 0.8289, "mean": 0.8292, "ev_only": 0.8279, "ref_ms": 8.2841, "cand_ms": 9.99181, "n_trials": 143, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 51.35, "compile_s": 2.62, "r_error": "", "src_len": 3232, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 3887, "ts": 1791533748.4770362, "chk": "ok", "code_len": 3232}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4046, "tout": 1805, "ts": 1791533876.0914285, "chk": null, "code_len": 3232}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791533714.994691}
{"ep": "h2-00027", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8081, "speedup": 0.8081, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 278.9, "calls": 2, "tokens_in": 14059, "tokens_out": 7276, "ctx_chars": 6513, "t_end": 1791533928.1450453, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8081, "lo95": 0.8068, "mean": 0.8063, "ev_only": 0.8079, "ref_ms": 7.18578, "cand_ms": 8.89216, "n_trials": 149, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0043, "r_wall_s": 206.5, "compile_s": 3.29, "r_error": "", "src_len": 5001, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 4833, "ts": 1791533692.3873687, "chk": "ok", "code_len": 5001}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8412, "tout": 2443, "ts": 1791533721.6020615, "chk": null, "code_len": 5001}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791533649.278546}
{"ep": "h2-00119", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.058, "speedup": 1.058, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 96.7, "calls": 3, "tokens_in": 22996, "tokens_out": 2465, "ctx_chars": 10527, "t_end": 1791533929.2704194, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.058, "lo95": 1.0569, "mean": 1.0592, "ev_only": 1.0572, "ref_ms": 19.67398, "cand_ms": 18.59544, "n_trials": 75, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0119, "r_wall_s": 60.71, "compile_s": 2.16, "r_error": "", "src_len": 2120, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1359, "ts": 1791533845.8739884, "chk": "ok", "code_len": 2335}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8132, "tout": 47, "ts": 1791533859.5495691, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8203, "tout": 1059, "ts": 1791533868.5217795, "chk": null, "code_len": 2120}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791533832.5829294}
{"ep": "h2-00071", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0129, "speedup": 1.0129, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 220.8, "calls": 2, "tokens_in": 5585, "tokens_out": 5616, "ctx_chars": 0, "t_end": 1791533931.9846532, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0129, "lo95": 1.0126, "mean": 1.0127, "ev_only": 1.0132, "ref_ms": 64.09969, "cand_ms": 63.28498, "n_trials": 24, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0049, "r_wall_s": 13.84, "compile_s": 1.96, "r_error": "", "src_len": 2828, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 4066, "ts": 1791533748.6128013, "chk": "ok", "code_len": 2828}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3839, "tout": 1550, "ts": 1791533918.1001534, "chk": null, "code_len": 2828}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791533711.1534843}
{"ep": "h2-00092", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7074, "speedup": 0.7074, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 190.7, "calls": 2, "tokens_in": 14266, "tokens_out": 1923, "ctx_chars": 10850, "t_end": 1791533942.3653123, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.7074, "lo95": 0.6794, "mean": 0.8779, "ev_only": 0.7143, "ref_ms": 0.17765, "cand_ms": 0.25113, "n_trials": 199, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 80.94, "compile_s": 3.33, "r_error": "", "src_len": 1215, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1261, "ts": 1791533763.1988292, "chk": "ok", "code_len": 1215}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7642, "tout": 662, "ts": 1791533861.148441, "chk": null, "code_len": 1215}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791533751.6788504}
{"ep": "h2-00080", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.6883, "speedup": 0.6883, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 211.7, "calls": 2, "tokens_in": 13860, "tokens_out": 13714, "ctx_chars": 6513, "t_end": 1791533942.3953354, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.6883, "lo95": 0.6845, "mean": 0.6867, "ev_only": 0.6859, "ref_ms": 15.94996, "cand_ms": 23.1719, "n_trials": 57, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.4158, "r_wall_s": 79.01, "compile_s": 3.08, "r_error": "", "src_len": 7252, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 10200, "ts": 1791533825.939844, "chk": "ok", "code_len": 7252}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8996, "tout": 3514, "ts": 1791533862.960218, "chk": null, "code_len": 7252}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791533730.7152917}
{"ep": "h2-00028", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.8835, "speedup": 2.8835, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 296.1, "calls": 2, "tokens_in": 7726, "tokens_out": 13049, "ctx_chars": 0, "t_end": 1791533945.4213667, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 2.8835, "lo95": 2.8554, "mean": 2.8694, "ev_only": 2.8826, "ref_ms": 60.22656, "cand_ms": 20.88675, "n_trials": 25, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8809, "r_wall_s": 25.59, "compile_s": 11.16, "r_error": "", "src_len": 6547, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 9834, "ts": 1791533735.4042416, "chk": "ok", "code_len": 6547}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5740, "tout": 3215, "ts": 1791533919.7117274, "chk": null, "code_len": 6547}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1986, "started": 1791533649.2787247}
{"ep": "h2-00042", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0287, "speedup": 1.0287, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 297.3, "calls": 2, "tokens_in": 14079, "tokens_out": 6272, "ctx_chars": 9266, "t_end": 1791533946.6146872, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0287, "lo95": 1.0285, "mean": 1.0288, "ev_only": 1.0289, "ref_ms": 30.43025, "cand_ms": 29.58072, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0158, "r_wall_s": 224.89, "compile_s": 3.28, "r_error": "", "src_len": 4569, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 3930, "ts": 1791533683.3366008, "chk": "ok", "code_len": 4569}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8463, "tout": 2342, "ts": 1791533721.264344, "chk": null, "code_len": 4569}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5616, "started": 1791533649.285249}
{"ep": "h2-00081", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0513, "speedup": 1.0513, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 215.5, "calls": 2, "tokens_in": 11432, "tokens_out": 3126, "ctx_chars": 6513, "t_end": 1791533946.6251063, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0513, "lo95": 1.051, "mean": 1.0712, "ev_only": 1.0521, "ref_ms": 15.71332, "cand_ms": 14.9462, "n_trials": 80, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.144, "r_wall_s": 30.33, "compile_s": 1.93, "r_error": "", "src_len": 3248, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1596, "ts": 1791533761.516594, "chk": "ok", "code_len": 3248}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6570, "tout": 1530, "ts": 1791533916.1220548, "chk": null, "code_len": 3248}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4862, "started": 1791533731.1741517}
{"ep": "h2-00079", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0427, "speedup": 1.0427, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 222.6, "calls": 2, "tokens_in": 13021, "tokens_out": 2326, "ctx_chars": 9274, "t_end": 1791533952.7556581, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0427, "lo95": 1.0422, "mean": 1.0412, "ev_only": 1.0435, "ref_ms": 8.29969, "cand_ms": 7.96017, "n_trials": 184, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0062, "r_wall_s": 26.12, "compile_s": 1.94, "r_error": "", "src_len": 1849, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1325, "ts": 1791533749.9351573, "chk": "ok", "code_len": 1849}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7229, "tout": 1001, "ts": 1791533926.4676063, "chk": null, "code_len": 1849}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791533730.1455452}
{"ep": "h2-00112", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0282, "speedup": 1.0282, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 138.6, "calls": 2, "tokens_in": 4489, "tokens_out": 2314, "ctx_chars": 0, "t_end": 1791533957.2120297, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0282, "lo95": 1.0271, "mean": 1.0282, "ev_only": 1.0283, "ref_ms": 24.94458, "cand_ms": 24.25963, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0227, "r_wall_s": 96.61, "compile_s": 3.5, "r_error": "", "src_len": 1096, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1644, "ts": 1791533832.9656441, "chk": "ok", "code_len": 1096}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2821, "tout": 670, "ts": 1791533859.441848, "chk": null, "code_len": 1096}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791533818.6600597}
{"ep": "h2-00137", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8457, "speedup": 0.8457, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 71.8, "calls": 1, "tokens_in": 7030, "tokens_out": 1211, "ctx_chars": 11175, "t_end": 1791533966.8116472, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8457, "lo95": 0.845, "mean": 0.8779, "ev_only": 0.8457, "ref_ms": 14.2438, "cand_ms": 16.84203, "n_trials": 73, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 61.04, "compile_s": 2.46, "r_error": "", "src_len": 2531, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1211, "ts": 1791533905.683939, "chk": null, "code_len": 2531}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791533894.966581}
{"ep": "h2-00122", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.326, "speedup": 1.326, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.5, "calls": 1, "tokens_in": 5998, "tokens_out": 733, "ctx_chars": 9970, "t_end": 1791533971.4549634, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.326, "lo95": 1.3251, "mean": 1.3224, "ev_only": 1.3271, "ref_ms": 7.88143, "cand_ms": 5.94387, "n_trials": 169, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0902, "r_wall_s": 117.91, "compile_s": 2.92, "r_error": "", "src_len": 1482, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 733, "ts": 1791533852.6147816, "chk": null, "code_len": 1482}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791533845.9161656}
{"ep": "h2-00004", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 6.768, "speedup": 6.768, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 334.8, "calls": 3, "tokens_in": 30360, "tokens_out": 24918, "ctx_chars": 8224, "t_end": 1791533984.1163528, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 6.768, "lo95": 6.2972, "mean": 7.0919, "ev_only": 6.7762, "ref_ms": 36.84485, "cand_ms": 5.444, "n_trials": 50, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9996, "r_wall_s": 47.03, "compile_s": 14.44, "r_error": "", "src_len": 7871, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 16023, "ts": 1791533792.7919743, "chk": "no_custom_kernel", "code_len": 7882}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 10214, "tout": 4760, "ts": 1791533845.7176347, "chk": "ok", "code_len": 7871}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 14636, "tout": 4135, "ts": 1791533937.0665412, "chk": null, "code_len": 7871}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791533649.2681158}
{"ep": "h2-00141", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0147, "speedup": 1.0147, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 78.5, "calls": 3, "tokens_in": 14997, "tokens_out": 2437, "ctx_chars": 5026, "t_end": 1791533985.1527202, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0147, "lo95": 1.0127, "mean": 1.0115, "ev_only": 1.0157, "ref_ms": 8.19486, "cand_ms": 8.07631, "n_trials": 179, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0113, "r_wall_s": 43.42, "compile_s": 2.02, "r_error": "", "src_len": 1946, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1341, "ts": 1791533919.6264257, "chk": "ok", "code_len": 1932}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5446, "tout": 88, "ts": 1791533933.3344405, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5558, "tout": 1008, "ts": 1791533941.6539893, "chk": null, "code_len": 1946}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 3993, "started": 1791533907.1759539}
{"ep": "h2-00138", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0123, "speedup": 1.0123, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.2, "calls": 2, "tokens_in": 4140, "tokens_out": 2292, "ctx_chars": 0, "t_end": 1791533992.8557513, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0123, "lo95": 1.0116, "mean": 1.0144, "ev_only": 1.0122, "ref_ms": 8.01077, "cand_ms": 7.91331, "n_trials": 169, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0054, "r_wall_s": 63.74, "compile_s": 2.14, "r_error": "", "src_len": 1150, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1584, "ts": 1791533909.2307906, "chk": "ok", "code_len": 1150}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2637, "tout": 708, "ts": 1791533929.0354712, "chk": null, "code_len": 1150}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791533895.699227}
{"ep": "h2-00145", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0171, "speedup": 1.0171, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 85.7, "calls": 3, "tokens_in": 25367, "tokens_out": 2610, "ctx_chars": 12352, "t_end": 1791533995.7101128, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0171, "lo95": 1.0163, "mean": 1.017, "ev_only": 1.0176, "ref_ms": 64.49687, "cand_ms": 63.41213, "n_trials": 19, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 34.97, "compile_s": 2.6, "r_error": "", "src_len": 2004, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1543, "ts": 1791533934.8896964, "chk": "ok", "code_len": 2140}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8983, "tout": 49, "ts": 1791533951.682187, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9056, "tout": 1018, "ts": 1791533960.692574, "chk": null, "code_len": 2004}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791533910.0009375}
{"ep": "h2-00139", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0047, "speedup": 1.0047, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 99.5, "calls": 1, "tokens_in": 6477, "tokens_out": 518, "ctx_chars": 10909, "t_end": 1791533996.403858, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0047, "lo95": 1.0044, "mean": 1.0304, "ev_only": 1.0053, "ref_ms": 7.97087, "cand_ms": 7.93368, "n_trials": 201, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 94.38, "compile_s": 2.05, "r_error": "", "src_len": 959, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 518, "ts": 1791533901.9527366, "chk": null, "code_len": 959}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791533896.8975654}
{"ep": "h2-00133", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0114, "speedup": 1.0114, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 118.9, "calls": 2, "tokens_in": 6441, "tokens_out": 7999, "ctx_chars": 0, "t_end": 1791533997.7841272, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0114, "lo95": 1.0108, "mean": 1.011, "ev_only": 1.0114, "ref_ms": 30.08942, "cand_ms": 29.74903, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0757, "r_wall_s": 42.41, "compile_s": 2.03, "r_error": "", "src_len": 5046, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 5298, "ts": 1791533923.9712784, "chk": "ok", "code_len": 5046}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4843, "tout": 2701, "ts": 1791533954.8418722, "chk": null, "code_len": 5046}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791533878.8827825}
{"ep": "h2-00047", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4637, "speedup": 1.4637, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 351.0, "calls": 2, "tokens_in": 15998, "tokens_out": 13821, "ctx_chars": 6513, "t_end": 1791534000.241417, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.4637, "lo95": 1.4587, "mean": 1.6465, "ev_only": 1.4677, "ref_ms": 3.29361, "cand_ms": 2.25012, "n_trials": 180, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0772, "r_wall_s": 128.62, "compile_s": 13.21, "r_error": "", "src_len": 9806, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 9265, "ts": 1791533729.5247023, "chk": "ok", "code_len": 9806}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10550, "tout": 4556, "ts": 1791533871.4823809, "chk": null, "code_len": 9806}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791533649.287039}
{"ep": "h2-00078", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0036, "speedup": 1.0036, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 270.3, "calls": 2, "tokens_in": 11408, "tokens_out": 1106, "ctx_chars": 9028, "t_end": 1791534000.2902095, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0036, "lo95": 1.0003, "mean": 1.0026, "ev_only": 1.0048, "ref_ms": 8.00471, "cand_ms": 7.97568, "n_trials": 144, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0083, "r_wall_s": 106.35, "compile_s": 2.36, "r_error": "", "src_len": 1024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 558, "ts": 1791533743.49151, "chk": "ok", "code_len": 1057}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6039, "tout": 548, "ts": 1791533893.8664725, "chk": null, "code_len": 1024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791533729.9851575}
{"ep": "h2-00149", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0005, "speedup": 1.0005, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 75.2, "calls": 2, "tokens_in": 11406, "tokens_out": 1121, "ctx_chars": 9028, "t_end": 1791534003.3388324, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0005, "lo95": 0.9996, "mean": 0.9997, "ev_only": 1.0015, "ref_ms": 7.98482, "cand_ms": 7.98092, "n_trials": 196, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0078, "r_wall_s": 55.44, "compile_s": 2.37, "r_error": "", "src_len": 1050, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 556, "ts": 1791533933.2938535, "chk": "ok", "code_len": 1037}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6037, "tout": 565, "ts": 1791533947.7460449, "chk": null, "code_len": 1050}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791533928.1453068}
{"ep": "h2-00021", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.188, "speedup": 1.188, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 354.8, "calls": 2, "tokens_in": 29024, "tokens_out": 18358, "ctx_chars": 0, "t_end": 1791534004.1043231, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.188, "lo95": 1.1811, "mean": 1.2335, "ev_only": 1.1854, "ref_ms": 19.15982, "cand_ms": 16.12758, "n_trials": 75, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1511, "r_wall_s": 31.8, "compile_s": 2.37, "r_error": "", "src_len": 16267, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 11040, "ts": 1791533734.4336221, "chk": "ok", "code_len": 16267}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18390, "tout": 7318, "ts": 1791533971.9309595, "chk": null, "code_len": 16267}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791533649.276132}
{"ep": "h2-00110", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0081, "speedup": 1.0081, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 195.4, "calls": 3, "tokens_in": 12687, "tokens_out": 5141, "ctx_chars": 0, "t_end": 1791534008.045378, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0081, "lo95": 1.0071, "mean": 1.007, "ev_only": 1.0079, "ref_ms": 64.39331, "cand_ms": 63.87868, "n_trials": 25, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0091, "r_wall_s": 135.13, "compile_s": 3.98, "r_error": "", "src_len": 3137, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1744, "tout": 3572, "ts": 1791533848.6825583, "chk": "ok", "code_len": 3172}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5428, "tout": 63, "ts": 1791533860.5054722, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5515, "tout": 1506, "ts": 1791533872.830869, "chk": null, "code_len": 3137}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1744, "started": 1791533812.6671476}
{"ep": "h2-00096", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4689, "speedup": 1.4689, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 253.6, "calls": 3, "tokens_in": 15863, "tokens_out": 12517, "ctx_chars": 0, "t_end": 1791534009.9373574, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.4689, "lo95": 1.4682, "mean": 1.4677, "ev_only": 1.4663, "ref_ms": 7.79678, "cand_ms": 5.30784, "n_trials": 182, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 75.06, "compile_s": 12.01, "r_error": "", "src_len": 4806, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6928, "ts": 1791533817.9474816, "chk": "no_custom_kernel", "code_len": 4860}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5442, "tout": 3001, "ts": 1791533864.9952724, "chk": "ok", "code_len": 4806}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8262, "tout": 2588, "ts": 1791533933.8768601, "chk": null, "code_len": 4806}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791533756.3403203}
{"ep": "h2-00118", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0406, "speedup": 1.0406, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 178.5, "calls": 2, "tokens_in": 5183, "tokens_out": 3210, "ctx_chars": 0, "t_end": 1791534010.6604135, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0406, "lo95": 1.0378, "mean": 1.042, "ev_only": 1.0433, "ref_ms": 8.50922, "cand_ms": 8.17759, "n_trials": 134, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0056, "r_wall_s": 135.7, "compile_s": 3.39, "r_error": "", "src_len": 1972, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 2082, "ts": 1791533850.2862246, "chk": "ok", "code_len": 1972}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3417, "tout": 1128, "ts": 1791533874.8171895, "chk": null, "code_len": 1972}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791533832.1551166}
{"ep": "h2-00053", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.6403, "speedup": 0.6403, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 364.1, "calls": 5, "tokens_in": 52462, "tokens_out": 12956, "ctx_chars": 0, "t_end": 1791534013.4218593, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.6403, "lo95": 0.6366, "mean": 0.6237, "ev_only": 0.639, "ref_ms": 7.26945, "cand_ms": 11.35278, "n_trials": 121, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0849, "r_wall_s": 40.3, "compile_s": 1.97, "r_error": "", "src_len": 3776, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2942, "tout": 5848, "ts": 1791533697.1532295, "chk": "incorrect", "code_len": 6940}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8922, "tout": 2271, "ts": 1791533741.4734743, "chk": "incorrect", "code_len": 3741}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 11325, "tout": 3141, "ts": 1791533938.8776414, "chk": "ok", "code_len": 3805}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 14575, "tout": 76, "ts": 1791533959.4003015, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 14698, "tout": 1620, "ts": 1791533973.0785723, "chk": null, "code_len": 3776}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "ok", "no_code", null], "first_tin": 2942, "started": 1791533649.28932}
{"ep": "h2-00130", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "incorrect", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 154.1, "calls": 1, "tokens_in": 2159, "tokens_out": 3029, "ctx_chars": 0, "t_end": 1791534022.245214, "error": "output mismatch beyond atol=rtol=0.0001", "f_status": "incorrect", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": 0.2512, "r_wall_s": 128.44, "compile_s": 12.54, "r_error": "output mismatch beyond atol=rtol=0.0001", "src_len": 6312, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2159, "tout": 3029, "ts": 1791533893.510296, "chk": null, "code_len": 6312}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2159, "started": 1791533868.0953994}
{"ep": "h2-00147", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0672, "speedup": 1.0672, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 104.9, "calls": 2, "tokens_in": 4887, "tokens_out": 3310, "ctx_chars": 0, "t_end": 1791534024.1332724, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0672, "lo95": 1.0554, "mean": 1.051, "ev_only": 0.9974, "ref_ms": 0.16327, "cand_ms": 0.15299, "n_trials": 164, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 55.74, "compile_s": 12.05, "r_error": "", "src_len": 2590, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 2023, "ts": 1791533935.998984, "chk": "ok", "code_len": 2590}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3238, "tout": 1287, "ts": 1791533968.1687927, "chk": null, "code_len": 2590}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791533919.246387}
{"ep": "h2-00135", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.5517, "speedup": 0.5517, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 145.6, "calls": 3, "tokens_in": 20134, "tokens_out": 3179, "ctx_chars": 9033, "t_end": 1791534025.567588, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.5517, "lo95": 0.5514, "mean": 0.5519, "ev_only": 0.5511, "ref_ms": 15.48572, "cand_ms": 28.07129, "n_trials": 50, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 51.8, "compile_s": 1.95, "r_error": "", "src_len": 2720, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1682, "ts": 1791533896.1389701, "chk": "ok", "code_len": 2720}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7279, "tout": 64, "ts": 1791533961.7289124, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7367, "tout": 1433, "ts": 1791533973.4341116, "chk": null, "code_len": 2720}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5488, "started": 1791533880.0114481}
{"ep": "h2-00088", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0267, "speedup": 1.0267, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 281.8, "calls": 2, "tokens_in": 4387, "tokens_out": 2282, "ctx_chars": 0, "t_end": 1791534026.679945, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0267, "lo95": 1.0245, "mean": 1.0257, "ev_only": 1.0273, "ref_ms": 24.9433, "cand_ms": 24.29499, "n_trials": 49, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0226, "r_wall_s": 84.96, "compile_s": 1.91, "r_error": "", "src_len": 1059, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1646, "ts": 1791533760.7483075, "chk": "ok", "code_len": 1059}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2719, "tout": 636, "ts": 1791533941.0166128, "chk": null, "code_len": 1059}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791533744.8946128}
{"ep": "h2-00082", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 27.1824, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 295.0, "calls": 2, "tokens_in": 7469, "tokens_out": 7726, "ctx_chars": 0, "t_end": 1791534026.7676365, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 27.1824, "lo95": 27.1411, "mean": 26.691, "ev_only": 27.7552, "ref_ms": 19.73909, "cand_ms": 0.72617, "n_trials": 72, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 86.65, "compile_s": 13.69, "r_error": "", "src_len": 5794, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 5004, "ts": 1791533778.2977512, "chk": "ok", "code_len": 5794}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5511, "tout": 2722, "ts": 1791533940.0769467, "chk": null, "code_len": 5794}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791533731.7248921}
{"ep": "h2-00116", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0092, "speedup": 1.0092, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 199.9, "calls": 2, "tokens_in": 5879, "tokens_out": 6671, "ctx_chars": 0, "t_end": 1791534027.580391, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0092, "lo95": 1.0083, "mean": 1.0091, "ev_only": 1.0085, "ref_ms": 30.12419, "cand_ms": 29.85031, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0195, "r_wall_s": 123.0, "compile_s": 2.1, "r_error": "", "src_len": 3787, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 4641, "ts": 1791533868.0025222, "chk": "ok", "code_len": 3787}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4281, "tout": 2030, "ts": 1791533903.8801587, "chk": null, "code_len": 3787}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791533827.7007265}
{"ep": "h2-00123", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.004, "speedup": 1.004, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 180.8, "calls": 2, "tokens_in": 11683, "tokens_out": 1403, "ctx_chars": 9028, "t_end": 1791534027.8136597, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.004, "lo95": 1.0034, "mean": 1.0044, "ev_only": 1.0044, "ref_ms": 8.00286, "cand_ms": 7.97135, "n_trials": 157, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 133.67, "compile_s": 2.44, "r_error": "", "src_len": 981, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 829, "ts": 1791533857.4768035, "chk": "ok", "code_len": 981}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6312, "tout": 574, "ts": 1791533894.073069, "chk": null, "code_len": 981}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791533846.9829242}
{"ep": "h2-00150", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9757, "speedup": 0.9757, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 99.1, "calls": 1, "tokens_in": 7032, "tokens_out": 1152, "ctx_chars": 11539, "t_end": 1791534028.4014952, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.9757, "lo95": 0.9752, "mean": 0.9749, "ev_only": 0.9759, "ref_ms": 31.42838, "cand_ms": 32.21094, "n_trials": 36, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0237, "r_wall_s": 85.56, "compile_s": 2.75, "r_error": "", "src_len": 1794, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1152, "ts": 1791533942.0255194, "chk": null, "code_len": 1794}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791533929.2706861}
{"ep": "h2-00076", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4824, "speedup": 1.4824, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 312.5, "calls": 2, "tokens_in": 6915, "tokens_out": 8215, "ctx_chars": 0, "t_end": 1791534028.7656894, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.4824, "lo95": 1.4822, "mean": 1.4821, "ev_only": 1.482, "ref_ms": 31.38458, "cand_ms": 21.17108, "n_trials": 52, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9573, "r_wall_s": 80.2, "compile_s": 2.39, "r_error": "", "src_len": 4025, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 6124, "ts": 1791533771.4576178, "chk": "ok", "code_len": 4025}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4837, "tout": 2091, "ts": 1791533947.6837108, "chk": null, "code_len": 4025}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791533716.2446582}
{"ep": "h2-00091", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.6426, "speedup": 0.6426, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 281.6, "calls": 5, "tokens_in": 51168, "tokens_out": 13754, "ctx_chars": 0, "t_end": 1791534030.065289, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.6426, "lo95": 0.6232, "mean": 0.6288, "ev_only": 0.6457, "ref_ms": 12.19574, "cand_ms": 18.97805, "n_trials": 61, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0912, "r_wall_s": 61.37, "compile_s": 1.96, "r_error": "", "src_len": 7246, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2942, "tout": 5783, "ts": 1791533796.3338325, "chk": "incorrect", "code_len": 7449}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8856, "tout": 1932, "ts": 1791533827.8445294, "chk": "ok", "code_len": 3705}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 10897, "tout": 3182, "ts": 1791533865.0495853, "chk": "ok", "code_len": 7246}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 14188, "tout": 50, "ts": 1791533946.162125, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 14285, "tout": 2807, "ts": 1791533968.441563, "chk": null, "code_len": 7246}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "ok", "no_code", null], "first_tin": 2942, "started": 1791533748.5032508}
{"ep": "h2-00126", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3305, "speedup": 1.3305, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 173.5, "calls": 1, "tokens_in": 5998, "tokens_out": 735, "ctx_chars": 9970, "t_end": 1791534031.2512033, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.3305, "lo95": 1.3293, "mean": 1.3305, "ev_only": 1.33, "ref_ms": 7.86376, "cand_ms": 5.91029, "n_trials": 173, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1111, "r_wall_s": 164.33, "compile_s": 3.87, "r_error": "", "src_len": 1484, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 735, "ts": 1791533866.0680358, "chk": null, "code_len": 1484}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791533857.7547328}
{"ep": "h2-00104", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0249, "speedup": 1.0249, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 247.3, "calls": 3, "tokens_in": 23576, "tokens_out": 7094, "ctx_chars": 9266, "t_end": 1791534036.0887265, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0249, "lo95": 1.0244, "mean": 1.0247, "ev_only": 1.0251, "ref_ms": 30.40142, "cand_ms": 29.6637, "n_trials": 39, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0165, "r_wall_s": 147.92, "compile_s": 2.39, "r_error": "", "src_len": 3548, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 3291, "ts": 1791533817.7288785, "chk": "forbidden", "code_len": 3521}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8001, "tout": 2016, "ts": 1791533832.6142364, "chk": "ok", "code_len": 3548}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9959, "tout": 1787, "ts": 1791533887.7635174, "chk": null, "code_len": 3548}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", null], "first_tin": 5616, "started": 1791533788.7700198}
{"ep": "h2-00005", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8741, "speedup": 0.8741, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 390.1, "calls": 3, "tokens_in": 64470, "tokens_out": 22035, "ctx_chars": 2425, "t_end": 1791534039.3414466, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8741, "lo95": 0.8704, "mean": 0.8737, "ev_only": 0.8741, "ref_ms": 19.7916, "cand_ms": 22.64286, "n_trials": 56, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.0299, "r_wall_s": 142.51, "compile_s": 12.66, "r_error": "", "src_len": 17255, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 14756, "ts": 1791533764.979282, "chk": "ok", "code_len": 17255}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 26413, "tout": 72, "ts": 1791533840.3703005, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 26509, "tout": 7207, "ts": 1791533896.767481, "chk": null, "code_len": 17255}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791533649.269227}
{"ep": "h2-00144", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8765, "speedup": 0.8765, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 131.8, "calls": 2, "tokens_in": 4583, "tokens_out": 2226, "ctx_chars": 0, "t_end": 1791534041.6933684, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.8765, "lo95": 0.8525, "mean": 0.8427, "ev_only": 0.9615, "ref_ms": 0.14662, "cand_ms": 0.16728, "n_trials": 185, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 81.66, "compile_s": 14.53, "r_error": "", "src_len": 2064, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 1176, "ts": 1791533928.7186995, "chk": "ok", "code_len": 2064}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2934, "tout": 1050, "ts": 1791533959.8388815, "chk": null, "code_len": 2064}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791533909.9208293}
{"ep": "h2-00143", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0018, "speedup": 1.0018, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 134.7, "calls": 3, "tokens_in": 24757, "tokens_out": 2255, "ctx_chars": 12117, "t_end": 1791534042.0839484, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0018, "lo95": 1.0004, "mean": 1.0016, "ev_only": 1.0018, "ref_ms": 42.22356, "cand_ms": 42.14634, "n_trials": 39, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 94.91, "compile_s": 1.98, "r_error": "", "src_len": 2287, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1131, "ts": 1791533917.3139315, "chk": "ok", "code_len": 2287}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8640, "tout": 56, "ts": 1791533937.3062313, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8720, "tout": 1068, "ts": 1791533946.3936446, "chk": null, "code_len": 2287}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791533907.354468}
{"ep": "h2-00121", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.8157, "speedup": 2.8157, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 200.2, "calls": 2, "tokens_in": 7579, "tokens_out": 11250, "ctx_chars": 0, "t_end": 1791534042.93297, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.8157, "lo95": 2.8087, "mean": 2.8247, "ev_only": 2.8411, "ref_ms": 14.38595, "cand_ms": 5.10927, "n_trials": 97, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9992, "r_wall_s": 84.13, "compile_s": 14.56, "r_error": "", "src_len": 5872, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 8678, "ts": 1791533922.457455, "chk": "ok", "code_len": 5872}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5396, "tout": 2572, "ts": 1791533958.6953597, "chk": null, "code_len": 5872}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2183, "started": 1791533843.2820876}
{"ep": "h2-00152", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3296, "speedup": 1.3296, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 108.0, "calls": 1, "tokens_in": 5998, "tokens_out": 718, "ctx_chars": 9970, "t_end": 1791534050.3249507, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.3296, "lo95": 1.3291, "mean": 1.3303, "ev_only": 1.3316, "ref_ms": 7.83026, "cand_ms": 5.88923, "n_trials": 201, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1101, "r_wall_s": 100.74, "compile_s": 2.62, "r_error": "", "src_len": 1433, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 718, "ts": 1791533948.9029994, "chk": null, "code_len": 1433}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791533942.3655856}
{"ep": "h2-00129", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9217, "speedup": 0.9217, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 184.8, "calls": 3, "tokens_in": 24018, "tokens_out": 2114, "ctx_chars": 11539, "t_end": 1791534051.884637, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9217, "lo95": 0.9213, "mean": 0.9221, "ev_only": 0.9225, "ref_ms": 7.76763, "cand_ms": 8.42719, "n_trials": 186, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0766, "r_wall_s": 79.51, "compile_s": 2.22, "r_error": "", "src_len": 1842, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1171, "ts": 1791533879.0674672, "chk": "ok", "code_len": 1842}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8393, "tout": 95, "ts": 1791533963.9336731, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8512, "tout": 848, "ts": 1791533971.4305017, "chk": null, "code_len": 1842}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791533867.0348725}
{"ep": "h2-00083", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.6076, "speedup": 1.6076, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 324.0, "calls": 3, "tokens_in": 15869, "tokens_out": 16102, "ctx_chars": 0, "t_end": 1791534057.6698227, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.6076, "lo95": 1.6067, "mean": 1.6075, "ev_only": 1.6097, "ref_ms": 42.90464, "cand_ms": 26.68863, "n_trials": 41, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9999, "r_wall_s": 92.11, "compile_s": 13.8, "r_error": "", "src_len": 5443, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 10049, "ts": 1791533837.9888499, "chk": "no_custom_kernel", "code_len": 5400}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5439, "tout": 3345, "ts": 1791533887.2960043, "chk": "ok", "code_len": 5443}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8394, "tout": 2708, "ts": 1791533964.834584, "chk": null, "code_len": 5443}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2036, "started": 1791533733.633433}
{"ep": "h2-00125", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8054, "speedup": 0.8054, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 203.6, "calls": 3, "tokens_in": 50101, "tokens_out": 14965, "ctx_chars": 2425, "t_end": 1791534057.7627254, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8054, "lo95": 0.7996, "mean": 0.8649, "ev_only": 0.8054, "ref_ms": 19.32897, "cand_ms": 23.99936, "n_trials": 35, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0297, "r_wall_s": 64.24, "compile_s": 1.93, "r_error": "", "src_len": 17539, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7592, "ts": 1791533923.5801034, "chk": "ok", "code_len": 17539}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19252, "tout": 25, "ts": 1791533936.5477893, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19301, "tout": 7348, "ts": 1791533993.454908, "chk": null, "code_len": 17539}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791533854.1836307}
{"ep": "h2-00094", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 6.1468, "speedup": 6.1468, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 312.1, "calls": 3, "tokens_in": 28178, "tokens_out": 17861, "ctx_chars": 8224, "t_end": 1791534066.1237001, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 6.1468, "lo95": 6.1324, "mean": 6.1596, "ev_only": 6.1408, "ref_ms": 31.78145, "cand_ms": 5.17042, "n_trials": 46, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9699, "r_wall_s": 27.96, "compile_s": 13.77, "r_error": "", "src_len": 6746, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 10420, "ts": 1791533845.3258734, "chk": "no_custom_kernel", "code_len": 6745}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9501, "tout": 4019, "ts": 1791533894.520379, "chk": "ok", "code_len": 6746}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 13167, "tout": 3422, "ts": 1791534038.1374145, "chk": null, "code_len": 6746}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791533754.0174525}
{"ep": "h2-00132", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.2994, "speedup": 0.2994, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 190.3, "calls": 3, "tokens_in": 21843, "tokens_out": 5202, "ctx_chars": 6513, "t_end": 1791534067.6573184, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.2994, "lo95": 0.249, "mean": 0.3828, "ev_only": 0.2981, "ref_ms": 3.94708, "cand_ms": 13.18166, "n_trials": 84, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0252, "r_wall_s": 110.21, "compile_s": 2.26, "r_error": "", "src_len": 5456, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2616, "ts": 1791533898.2001183, "chk": "ok", "code_len": 5456}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8174, "tout": 25, "ts": 1791533937.2929194, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8223, "tout": 2561, "ts": 1791533957.422508, "chk": null, "code_len": 5456}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791533877.317507}
{"ep": "h2-00111", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.0184, "speedup": 3.0184, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.2, "calls": 2, "tokens_in": 6181, "tokens_out": 8026, "ctx_chars": 0, "t_end": 1791534068.3269534, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 3.0184, "lo95": 3.0165, "mean": 3.0161, "ev_only": 3.0174, "ref_ms": 27.65286, "cand_ms": 9.16154, "n_trials": 45, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.8948, "r_wall_s": 26.55, "compile_s": 2.85, "r_error": "", "src_len": 3902, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 5944, "ts": 1791533868.9368706, "chk": "ok", "code_len": 3902}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4451, "tout": 2082, "ts": 1791534041.7508416, "chk": null, "code_len": 3902}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791533816.141101}
{"ep": "h2-00188", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0056, "speedup": 1.0056, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 27.8, "calls": 1, "tokens_in": 6477, "tokens_out": 490, "ctx_chars": 10909, "t_end": 1791534069.527296, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0056, "lo95": 1.0052, "mean": 1.006, "ev_only": 1.0053, "ref_ms": 7.98421, "cand_ms": 7.94011, "n_trials": 162, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0043, "r_wall_s": 22.35, "compile_s": 2.67, "r_error": "", "src_len": 932, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 490, "ts": 1791534047.01731, "chk": null, "code_len": 932}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791534041.6936185}
{"ep": "h2-00134", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.1219, "speedup": 0.1219, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 191.3, "calls": 3, "tokens_in": 10439, "tokens_out": 3617, "ctx_chars": 0, "t_end": 1791534070.2191768, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.1219, "lo95": 0.1218, "mean": 0.1217, "ev_only": 0.1219, "ref_ms": 42.69288, "cand_ms": 350.16703, "n_trials": 13, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 101.8, "compile_s": 1.94, "r_error": "", "src_len": 2876, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2034, "tout": 1810, "ts": 1791533898.0824957, "chk": "ok", "code_len": 2994}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3956, "tout": 469, "ts": 1791533956.3198533, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 4449, "tout": 1338, "ts": 1791533967.4805703, "chk": null, "code_len": 2876}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 2034, "started": 1791533878.8921509}
{"ep": "h2-00124", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.6597, "speedup": 0.6597, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 220.9, "calls": 3, "tokens_in": 18683, "tokens_out": 1469, "ctx_chars": 8224, "t_end": 1791534070.382136, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.6597, "lo95": 0.6165, "mean": 0.72, "ev_only": 0.6598, "ref_ms": 33.09883, "cand_ms": 50.16996, "n_trials": 31, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 88.53, "compile_s": 2.28, "r_error": "", "src_len": 1005, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 931, "ts": 1791533869.8229346, "chk": "ok", "code_len": 1005}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6548, "tout": 55, "ts": 1791533974.9129608, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6627, "tout": 483, "ts": 1791533981.806909, "chk": null, "code_len": 1005}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791533849.5311894}
{"ep": "h2-00114", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4874, "speedup": 1.4874, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 248.4, "calls": 2, "tokens_in": 30358, "tokens_out": 16323, "ctx_chars": 2425, "t_end": 1791534070.6416333, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.4874, "lo95": 1.4813, "mean": 1.5462, "ev_only": 1.4904, "ref_ms": 19.15141, "cand_ms": 12.87542, "n_trials": 51, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1091, "r_wall_s": 37.12, "compile_s": 2.32, "r_error": "", "src_len": 15440, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 9643, "ts": 1791533896.4560084, "chk": "ok", "code_len": 15482}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18808, "tout": 6680, "ts": 1791534033.4853613, "chk": null, "code_len": 15440}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791533822.217695}
{"ep": "h2-00177", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0069, "speedup": 1.0069, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 45.1, "calls": 2, "tokens_in": 11683, "tokens_out": 1407, "ctx_chars": 9028, "t_end": 1791534070.6646059, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0069, "lo95": 1.0066, "mean": 1.0068, "ev_only": 1.006, "ref_ms": 7.98097, "cand_ms": 7.92642, "n_trials": 161, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0043, "r_wall_s": 22.78, "compile_s": 2.72, "r_error": "", "src_len": 991, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 829, "ts": 1791534032.38828, "chk": "ok", "code_len": 991}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6312, "tout": 578, "ts": 1791534047.7037854, "chk": null, "code_len": 991}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791534025.5678668}
{"ep": "h2-00175", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0052, "speedup": 1.0052, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 50.8, "calls": 2, "tokens_in": 13696, "tokens_out": 1158, "ctx_chars": 10909, "t_end": 1791534073.0744648, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0052, "lo95": 1.0019, "mean": 1.0052, "ev_only": 1.0064, "ref_ms": 8.01803, "cand_ms": 7.97617, "n_trials": 154, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 27.22, "compile_s": 2.95, "r_error": "", "src_len": 968, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 629, "ts": 1791534028.281401, "chk": "ok", "code_len": 968}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7217, "tout": 529, "ts": 1791534045.6966763, "chk": null, "code_len": 968}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791534022.245502}
{"ep": "h2-00067", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0216, "speedup": 1.0216, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 372.4, "calls": 3, "tokens_in": 13840, "tokens_out": 9818, "ctx_chars": 0, "t_end": 1791534074.0928867, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0216, "lo95": 1.0209, "mean": 1.0061, "ev_only": 1.0213, "ref_ms": 30.4895, "cand_ms": 29.84358, "n_trials": 48, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0152, "r_wall_s": 120.86, "compile_s": 1.99, "r_error": "", "src_len": 5103, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 4408, "ts": 1791533738.1678424, "chk": "forbidden", "code_len": 4918}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4669, "tout": 2765, "ts": 1791533758.2380583, "chk": "ok", "code_len": 5103}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7546, "tout": 2645, "ts": 1791533952.7310483, "chk": null, "code_len": 5103}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", null], "first_tin": 1625, "started": 1791533701.6446383}
{"ep": "h2-00128", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0435, "speedup": 1.0435, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 210.8, "calls": 2, "tokens_in": 12746, "tokens_out": 1951, "ctx_chars": 9274, "t_end": 1791534075.902323, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0435, "lo95": 1.0432, "mean": 1.0433, "ev_only": 1.0444, "ref_ms": 8.28857, "cand_ms": 7.94298, "n_trials": 139, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0045, "r_wall_s": 33.67, "compile_s": 2.16, "r_error": "", "src_len": 1610, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1050, "ts": 1791533873.7890267, "chk": "ok", "code_len": 1610}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6954, "tout": 901, "ts": 1791534042.1607325, "chk": null, "code_len": 1610}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791533865.079269}
{"ep": "h2-00153", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0282, "speedup": 1.0282, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 135.2, "calls": 1, "tokens_in": 5832, "tokens_out": 542, "ctx_chars": 9656, "t_end": 1791534077.5854049, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0282, "lo95": 1.0266, "mean": 1.0275, "ev_only": 1.0277, "ref_ms": 24.85326, "cand_ms": 24.17273, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0224, "r_wall_s": 129.02, "compile_s": 2.14, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 542, "ts": 1791533947.834538, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791533942.3955412}
{"ep": "h2-00186", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.011, "speedup": 1.011, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 44.6, "calls": 1, "tokens_in": 1599, "tokens_out": 594, "ctx_chars": 0, "t_end": 1791534080.6674793, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.011, "lo95": 1.0104, "mean": 1.0113, "ev_only": 1.0103, "ref_ms": 7.98532, "cand_ms": 7.89871, "n_trials": 196, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0052, "r_wall_s": 38.16, "compile_s": 2.16, "r_error": "", "src_len": 1061, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1599, "tout": 594, "ts": 1791534042.444147, "chk": null, "code_len": 1061}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1599, "started": 1791534036.0890174}
{"ep": "h2-00155", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.016, "speedup": 1.016, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 142.7, "calls": 3, "tokens_in": 24697, "tokens_out": 2177, "ctx_chars": 12117, "t_end": 1791534089.3391457, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.016, "lo95": 1.0151, "mean": 1.0169, "ev_only": 1.0156, "ref_ms": 42.71274, "cand_ms": 42.03862, "n_trials": 30, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 96.49, "compile_s": 1.89, "r_error": "", "src_len": 2233, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1116, "ts": 1791533957.6224654, "chk": "ok", "code_len": 2233}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8625, "tout": 26, "ts": 1791533982.9687269, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8675, "tout": 1035, "ts": 1791533991.8030062, "chk": null, "code_len": 2233}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791533946.6149533}
{"ep": "h2-00154", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0059, "speedup": 1.0059, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 144.7, "calls": 3, "tokens_in": 19851, "tokens_out": 2135, "ctx_chars": 9274, "t_end": 1791534090.1034703, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0059, "lo95": 1.0046, "mean": 1.0055, "ev_only": 1.0055, "ref_ms": 8.32948, "cand_ms": 8.28099, "n_trials": 150, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0448, "r_wall_s": 81.51, "compile_s": 2.04, "r_error": "", "src_len": 1955, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1089, "ts": 1791533955.887509, "chk": "ok", "code_len": 2008}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6991, "tout": 55, "ts": 1791533999.9448664, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7070, "tout": 991, "ts": 1791534008.3017418, "chk": null, "code_len": 1955}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791533945.4216092}
{"ep": "h2-00085", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5826, "speedup": 1.5826, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 366.3, "calls": 3, "tokens_in": 15965, "tokens_out": 12873, "ctx_chars": 0, "t_end": 1791534106.459955, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.5826, "lo95": 1.5816, "mean": 1.5819, "ev_only": 1.5833, "ref_ms": 7.77154, "cand_ms": 4.91064, "n_trials": 199, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 128.27, "compile_s": 12.51, "r_error": "", "src_len": 5412, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6981, "ts": 1791533802.250858, "chk": "no_custom_kernel", "code_len": 5415}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5463, "tout": 3166, "ts": 1791533851.490535, "chk": "ok", "code_len": 5412}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8343, "tout": 2726, "ts": 1791533977.507429, "chk": null, "code_len": 5412}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791533740.1604068}
{"ep": "h2-00159", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0212, "speedup": 1.0212, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 140.1, "calls": 2, "tokens_in": 9215, "tokens_out": 2504, "ctx_chars": 5026, "t_end": 1791534106.930869, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0212, "lo95": 1.0205, "mean": 1.02, "ev_only": 1.0238, "ref_ms": 8.18306, "cand_ms": 8.01347, "n_trials": 201, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0086, "r_wall_s": 39.37, "compile_s": 2.06, "r_error": "", "src_len": 1465, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1706, "ts": 1791534000.8781586, "chk": "ok", "code_len": 1465}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5220, "tout": 798, "ts": 1791534067.461651, "chk": null, "code_len": 1465}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791533967.3456283}
{"ep": "h2-00156", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9995, "speedup": 0.9995, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 162.9, "calls": 3, "tokens_in": 11412, "tokens_out": 3854, "ctx_chars": 0, "t_end": 1791534109.5202253, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9995, "lo95": 0.9837, "mean": 0.9968, "ev_only": 0.9997, "ref_ms": 1.79926, "cand_ms": 1.8001, "n_trials": 223, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.2915, "r_wall_s": 112.28, "compile_s": 2.04, "r_error": "", "src_len": 1865, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1757, "tout": 2940, "ts": 1791533977.5145652, "chk": "ok", "code_len": 1883}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4809, "tout": 13, "ts": 1791533989.3562858, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 4846, "tout": 901, "ts": 1791533997.1858225, "chk": null, "code_len": 1865}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1757, "started": 1791533946.6253138}
{"ep": "h2-00194", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8629, "speedup": 0.8629, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 54.2, "calls": 1, "tokens_in": 1647, "tokens_out": 434, "ctx_chars": 0, "t_end": 1791534111.939261, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8629, "lo95": 0.8278, "mean": 0.8142, "ev_only": 0.8047, "ref_ms": 0.18624, "cand_ms": 0.21582, "n_trials": 201, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 49.21, "compile_s": 1.89, "r_error": "", "src_len": 782, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1647, "tout": 434, "ts": 1791534062.6075234, "chk": null, "code_len": 782}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1647, "started": 1791534057.7629688}
{"ep": "h2-00178", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0043, "speedup": 1.0043, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 88.2, "calls": 3, "tokens_in": 20093, "tokens_out": 2780, "ctx_chars": 9266, "t_end": 1791534114.9255865, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0043, "lo95": 1.0035, "mean": 1.0039, "ev_only": 1.0046, "ref_ms": 30.55923, "cand_ms": 30.42796, "n_trials": 41, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.041, "r_wall_s": 43.15, "compile_s": 2.06, "r_error": "", "src_len": 2623, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1489, "ts": 1791534039.7940483, "chk": "ok", "code_len": 2623}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7215, "tout": 25, "ts": 1791534060.4822965, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7264, "tout": 1266, "ts": 1791534071.3586211, "chk": null, "code_len": 2623}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791534026.6801958}
{"ep": "h2-00157", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0077, "speedup": 1.0077, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 163.9, "calls": 2, "tokens_in": 16164, "tokens_out": 2369, "ctx_chars": 12117, "t_end": 1791534116.6848114, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0077, "lo95": 1.0067, "mean": 1.0078, "ev_only": 1.0076, "ref_ms": 42.46586, "cand_ms": 42.14214, "n_trials": 39, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 91.81, "compile_s": 2.89, "r_error": "", "src_len": 2292, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1258, "ts": 1791533964.5279665, "chk": "ok", "code_len": 2292}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8767, "tout": 1111, "ts": 1791534023.720535, "chk": null, "code_len": 2292}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7397, "started": 1791533952.7559292}
{"ep": "h2-00140", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4376, "speedup": 1.4376, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 220.3, "calls": 2, "tokens_in": 12015, "tokens_out": 8416, "ctx_chars": 6513, "t_end": 1791534118.8401976, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.4376, "lo95": 1.4362, "mean": 1.4228, "ev_only": 1.439, "ref_ms": 15.48082, "cand_ms": 10.76824, "n_trials": 105, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2113, "r_wall_s": 87.52, "compile_s": 2.02, "r_error": "", "src_len": 4955, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 5934, "ts": 1791533951.2837899, "chk": "ok", "code_len": 4955}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7408, "tout": 2482, "ts": 1791534030.419589, "chk": null, "code_len": 4955}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791533898.5196848}
{"ep": "h2-00171", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8455, "speedup": 0.8455, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 112.0, "calls": 2, "tokens_in": 15743, "tokens_out": 3118, "ctx_chars": 11175, "t_end": 1791534120.0491502, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8455, "lo95": 0.8446, "mean": 0.8478, "ev_only": 0.8471, "ref_ms": 14.28786, "cand_ms": 16.89924, "n_trials": 66, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 63.52, "compile_s": 2.08, "r_error": "", "src_len": 2470, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1829, "ts": 1791534024.6445742, "chk": "ok", "code_len": 2470}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8711, "tout": 1289, "ts": 1791534056.4433634, "chk": null, "code_len": 2470}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791534008.0456371}
{"ep": "h2-00120", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.8224, "speedup": 2.8224, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 284.7, "calls": 2, "tokens_in": 6362, "tokens_out": 7059, "ctx_chars": 0, "t_end": 1791534121.941968, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.8224, "lo95": 2.8204, "mean": 2.8186, "ev_only": 2.8194, "ref_ms": 8.50871, "cand_ms": 3.01473, "n_trials": 176, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.3038, "r_wall_s": 123.73, "compile_s": 12.64, "r_error": "", "src_len": 4536, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 4754, "ts": 1791533878.5320537, "chk": "ok", "code_len": 4536}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4542, "tout": 2305, "ts": 1791533998.1372042, "chk": null, "code_len": 4536}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791533837.2191107}
{"ep": "h2-00136", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 5.024, "speedup": 5.024, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 228.5, "calls": 2, "tokens_in": 6325, "tokens_out": 6718, "ctx_chars": 0, "t_end": 1791534122.7262993, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 5.024, "lo95": 4.9644, "mean": 4.7762, "ev_only": 5.165, "ref_ms": 1.8543, "cand_ms": 0.36909, "n_trials": 225, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 153.95, "compile_s": 12.34, "r_error": "", "src_len": 4643, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4390, "ts": 1791533935.5520651, "chk": "ok", "code_len": 4643}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4566, "tout": 2328, "ts": 1791533968.7289393, "chk": null, "code_len": 4643}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1759, "started": 1791533894.2105293}
{"ep": "h2-00151", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0231, "speedup": 1.0231, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 191.0, "calls": 2, "tokens_in": 13648, "tokens_out": 5100, "ctx_chars": 9266, "t_end": 1791534122.9659188, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0231, "lo95": 1.0225, "mean": 1.023, "ev_only": 1.0232, "ref_ms": 30.47942, "cand_ms": 29.79265, "n_trials": 40, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0203, "r_wall_s": 104.09, "compile_s": 3.08, "r_error": "", "src_len": 3704, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 3234, "ts": 1791533959.7631936, "chk": "ok", "code_len": 3759}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8032, "tout": 1866, "ts": 1791534018.342902, "chk": null, "code_len": 3704}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5616, "started": 1791533931.9849563}
{"ep": "h2-00117", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.968, "speedup": 0.968, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 303.0, "calls": 2, "tokens_in": 12362, "tokens_out": 20091, "ctx_chars": 0, "t_end": 1791534132.1208863, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.968, "lo95": 0.9672, "mean": 0.9702, "ev_only": 0.9665, "ref_ms": 7.09757, "cand_ms": 7.33215, "n_trials": 174, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.997, "r_wall_s": 53.26, "compile_s": 1.98, "r_error": "", "src_len": 11666, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 14160, "ts": 1791533949.1207798, "chk": "ok", "code_len": 11666}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9418, "tout": 5931, "ts": 1791534078.8348553, "chk": null, "code_len": 11666}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791533829.1421125}
{"ep": "h2-00184", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.3777, "speedup": 2.3777, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 102.4, "calls": 2, "tokens_in": 15194, "tokens_out": 5831, "ctx_chars": 8529, "t_end": 1791534132.4521425, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.3777, "lo95": 2.3756, "mean": 2.3968, "ev_only": 2.3808, "ref_ms": 8.52498, "cand_ms": 3.58541, "n_trials": 176, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4108, "r_wall_s": 31.71, "compile_s": 1.95, "r_error": "", "src_len": 3344, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 4092, "ts": 1791534065.9126446, "chk": "ok", "code_len": 3344}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9699, "tout": 1739, "ts": 1791534100.6774807, "chk": null, "code_len": 3344}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791534030.0655613}
{"ep": "h2-00176", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7117, "speedup": 0.7117, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 110.4, "calls": 2, "tokens_in": 16440, "tokens_out": 4416, "ctx_chars": 12352, "t_end": 1791534134.579805, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.7117, "lo95": 0.7112, "mean": 0.7122, "ev_only": 0.7115, "ref_ms": 64.29093, "cand_ms": 90.3387, "n_trials": 16, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0019, "r_wall_s": 61.98, "compile_s": 2.3, "r_error": "", "src_len": 2208, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 3233, "ts": 1791534053.7797945, "chk": "ok", "code_len": 2208}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9110, "tout": 1183, "ts": 1791534072.5772643, "chk": null, "code_len": 2208}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791534024.1335342}
{"ep": "h2-00127", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.2938, "speedup": 1.2938, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 277.2, "calls": 2, "tokens_in": 30768, "tokens_out": 17522, "ctx_chars": 2425, "t_end": 1791534139.0642288, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.2938, "lo95": 1.2687, "mean": 1.2958, "ev_only": 1.301, "ref_ms": 19.51383, "cand_ms": 15.08218, "n_trials": 55, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1477, "r_wall_s": 128.37, "compile_s": 2.0, "r_error": "", "src_len": 16376, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 10275, "ts": 1791533941.1597815, "chk": "ok", "code_len": 16376}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19218, "tout": 7247, "ts": 1791534010.3160129, "chk": null, "code_len": 16376}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791533861.9036741}
{"ep": "h2-00146", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.6654, "speedup": 1.6654, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 235.5, "calls": 2, "tokens_in": 7412, "tokens_out": 8306, "ctx_chars": 0, "t_end": 1791534146.2023764, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.6654, "lo95": 1.665, "mean": 1.6647, "ev_only": 1.6629, "ref_ms": 7.75624, "cand_ms": 4.65715, "n_trials": 163, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 101.97, "compile_s": 14.24, "r_error": "", "src_len": 5085, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 5752, "ts": 1791533966.3398223, "chk": "ok", "code_len": 5085}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5253, "tout": 2554, "ts": 1791534043.5152318, "chk": null, "code_len": 5085}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2159, "started": 1791533910.6765947}
{"ep": "h2-00206", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0451, "speedup": 1.0451, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 70.5, "calls": 2, "tokens_in": 4997, "tokens_out": 2905, "ctx_chars": 0, "t_end": 1791534148.0890687, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0451, "lo95": 1.0442, "mean": 1.0435, "ev_only": 1.047, "ref_ms": 8.34606, "cand_ms": 7.98605, "n_trials": 147, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0049, "r_wall_s": 37.56, "compile_s": 2.12, "r_error": "", "src_len": 1842, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 1946, "ts": 1791534094.0002391, "chk": "ok", "code_len": 1842}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3231, "tout": 959, "ts": 1791534110.4498935, "chk": null, "code_len": 1842}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791534077.5856788}
{"ep": "h2-00158", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0222, "speedup": 1.0222, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 191.8, "calls": 2, "tokens_in": 6003, "tokens_out": 6175, "ctx_chars": 0, "t_end": 1791534149.038913, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0222, "lo95": 1.0218, "mean": 1.0485, "ev_only": 1.0228, "ref_ms": 30.36952, "cand_ms": 29.71046, "n_trials": 52, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0196, "r_wall_s": 81.94, "compile_s": 2.05, "r_error": "", "src_len": 3633, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 3988, "ts": 1791533991.8515952, "chk": "ok", "code_len": 4297}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4378, "tout": 2187, "ts": 1791534066.2805862, "chk": null, "code_len": 3633}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791533957.2122972}
{"ep": "h2-00192", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0545, "speedup": 1.0545, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 101.6, "calls": 2, "tokens_in": 13891, "tokens_out": 3492, "ctx_chars": 10167, "t_end": 1791534153.515171, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0545, "lo95": 1.05, "mean": 1.0582, "ev_only": 1.0567, "ref_ms": 8.56627, "cand_ms": 8.12375, "n_trials": 170, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0069, "r_wall_s": 46.59, "compile_s": 1.94, "r_error": "", "src_len": 2382, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 2279, "ts": 1791534071.5229554, "chk": "ok", "code_len": 2382}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7785, "tout": 1213, "ts": 1791534106.8368847, "chk": null, "code_len": 2382}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791534051.8849254}
{"ep": "h2-00190", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.8958, "speedup": 1.8958, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 112.4, "calls": 2, "tokens_in": 15169, "tokens_out": 4910, "ctx_chars": 11065, "t_end": 1791534155.3623838, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.8958, "lo95": 1.8807, "mean": 1.9094, "ev_only": 1.9589, "ref_ms": 1.85071, "cand_ms": 0.97621, "n_trials": 192, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 36.31, "compile_s": 2.06, "r_error": "", "src_len": 3044, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 3415, "ts": 1791534074.8768365, "chk": "ok", "code_len": 3044}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8540, "tout": 1495, "ts": 1791534119.0070338, "chk": null, "code_len": 3044}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791534042.9332254}
{"ep": "h2-00165", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9685, "speedup": 0.9685, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 160.5, "calls": 2, "tokens_in": 12085, "tokens_out": 2583, "ctx_chars": 9033, "t_end": 1791534156.868565, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9685, "lo95": 0.968, "mean": 0.9704, "ev_only": 0.9688, "ref_ms": 15.51438, "cand_ms": 16.01966, "n_trials": 86, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0194, "r_wall_s": 58.35, "compile_s": 1.94, "r_error": "", "src_len": 1091, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 1904, "ts": 1791534014.5551589, "chk": "ok", "code_len": 1091}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6595, "tout": 679, "ts": 1791534097.9983318, "chk": null, "code_len": 1091}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791533996.4041069}
{"ep": "h2-00172", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0117, "speedup": 1.0117, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.3, "calls": 3, "tokens_in": 23721, "tokens_out": 3539, "ctx_chars": 10167, "t_end": 1791534164.2242534, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0117, "lo95": 1.0109, "mean": 1.0118, "ev_only": 1.0112, "ref_ms": 8.47071, "cand_ms": 8.3727, "n_trials": 150, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0557, "r_wall_s": 102.53, "compile_s": 2.0, "r_error": "", "src_len": 1900, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 2550, "ts": 1791534037.999579, "chk": "ok", "code_len": 1900}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8766, "tout": 61, "ts": 1791534053.4717479, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8851, "tout": 928, "ts": 1791534061.621557, "chk": null, "code_len": 1900}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791534009.9376237}
{"ep": "h2-00196", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9786, "speedup": 0.9786, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.8, "calls": 2, "tokens_in": 12245, "tokens_out": 2857, "ctx_chars": 9033, "t_end": 1791534165.4091477, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9786, "lo95": 0.9782, "mean": 0.9784, "ev_only": 0.9793, "ref_ms": 15.53193, "cand_ms": 15.87234, "n_trials": 84, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0148, "r_wall_s": 38.9, "compile_s": 2.04, "r_error": "", "src_len": 1222, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 2070, "ts": 1791534087.905283, "chk": "ok", "code_len": 1222}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6755, "tout": 787, "ts": 1791534126.075506, "chk": null, "code_len": 1222}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791534067.6575856}
{"ep": "h2-00148", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0079, "speedup": 1.0079, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 238.1, "calls": 3, "tokens_in": 19983, "tokens_out": 2693, "ctx_chars": 9266, "t_end": 1791534165.61285, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0079, "lo95": 1.0069, "mean": 1.0072, "ev_only": 1.0082, "ref_ms": 30.51668, "cand_ms": 30.27815, "n_trials": 40, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0414, "r_wall_s": 192.3, "compile_s": 2.58, "r_error": "", "src_len": 2457, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1417, "ts": 1791533941.0359783, "chk": "ok", "code_len": 2457}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7143, "tout": 59, "ts": 1791533962.4683197, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7226, "tout": 1217, "ts": 1791533972.5008507, "chk": null, "code_len": 2457}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791533927.5182061}
{"ep": "h2-00168", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8024, "speedup": 0.8024, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 166.0, "calls": 1, "tokens_in": 1902, "tokens_out": 2081, "ctx_chars": 0, "t_end": 1791534166.3385444, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8024, "lo95": 0.8019, "mean": 0.8025, "ev_only": 0.8018, "ref_ms": 15.52622, "cand_ms": 19.34999, "n_trials": 76, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.5041, "r_wall_s": 144.51, "compile_s": 56.19, "r_error": "", "src_len": 3772, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1902, "tout": 2081, "ts": 1791534020.8525736, "chk": null, "code_len": 3772}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1902, "started": 1791534000.2904143}
{"ep": "h2-00162", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0297, "speedup": 1.0297, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 183.2, "calls": 2, "tokens_in": 9284, "tokens_out": 2690, "ctx_chars": 5026, "t_end": 1791534168.3104527, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0297, "lo95": 1.0289, "mean": 1.03, "ev_only": 1.0285, "ref_ms": 8.14906, "cand_ms": 7.91434, "n_trials": 182, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.008, "r_wall_s": 44.43, "compile_s": 2.39, "r_error": "", "src_len": 1427, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1932, "ts": 1791534003.4559581, "chk": "ok", "code_len": 1427}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5289, "tout": 758, "ts": 1791534123.7805982, "chk": null, "code_len": 1427}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791533985.153006}
{"ep": "h2-00185", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5972, "speedup": 1.5972, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 140.4, "calls": 2, "tokens_in": 5715, "tokens_out": 4621, "ctx_chars": 0, "t_end": 1791534171.6375625, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.5972, "lo95": 1.5948, "mean": 1.5912, "ev_only": 1.5928, "ref_ms": 8.18716, "cand_ms": 5.12594, "n_trials": 166, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 61.37, "compile_s": 2.52, "r_error": "", "src_len": 2897, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 3015, "ts": 1791534056.4268582, "chk": "ok", "code_len": 2897}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3963, "tout": 1606, "ts": 1791534110.1648467, "chk": null, "code_len": 2897}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1752, "started": 1791534031.2514808}
{"ep": "h2-00164", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0072, "speedup": 1.0072, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 179.5, "calls": 2, "tokens_in": 4080, "tokens_out": 1886, "ctx_chars": 0, "t_end": 1791534175.1978574, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0072, "lo95": 1.006, "mean": 1.0076, "ev_only": 1.0089, "ref_ms": 7.98081, "cand_ms": 7.92414, "n_trials": 89, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 66.88, "compile_s": 2.66, "r_error": "", "src_len": 960, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 1309, "ts": 1791534008.1687367, "chk": "ok", "code_len": 960}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2456, "tout": 577, "ts": 1791534108.230432, "chk": null, "code_len": 960}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791533995.7103744}
{"ep": "h2-00169", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0254, "speedup": 1.0254, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 172.8, "calls": 2, "tokens_in": 12581, "tokens_out": 1715, "ctx_chars": 9656, "t_end": 1791534176.13709, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0254, "lo95": 1.0241, "mean": 1.0255, "ev_only": 1.0256, "ref_ms": 24.87488, "cand_ms": 24.25789, "n_trials": 44, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0287, "r_wall_s": 70.94, "compile_s": 1.86, "r_error": "", "src_len": 1031, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 1099, "ts": 1791534013.425328, "chk": "ok", "code_len": 1031}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6747, "tout": 616, "ts": 1791534104.072484, "chk": null, "code_len": 1031}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791534003.3391068}
{"ep": "h2-00161", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.44, "speedup": 1.44, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 195.3, "calls": 2, "tokens_in": 12035, "tokens_out": 8065, "ctx_chars": 6513, "t_end": 1791534179.426407, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.44, "lo95": 1.4392, "mean": 1.441, "ev_only": 1.4385, "ref_ms": 15.47077, "cand_ms": 10.74371, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2745, "r_wall_s": 107.09, "compile_s": 1.94, "r_error": "", "src_len": 4836, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 5778, "ts": 1791534035.4674065, "chk": "ok", "code_len": 4836}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7428, "tout": 2287, "ts": 1791534071.670268, "chk": null, "code_len": 4836}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791533984.116666}
{"ep": "h2-00180", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.004, "speedup": 1.004, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 152.6, "calls": 3, "tokens_in": 24932, "tokens_out": 2339, "ctx_chars": 12117, "t_end": 1791534180.1519277, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.004, "lo95": 1.0028, "mean": 1.0035, "ev_only": 1.0047, "ref_ms": 42.25204, "cand_ms": 42.08462, "n_trials": 35, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 110.99, "compile_s": 2.05, "r_error": "", "src_len": 2267, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1221, "ts": 1791534039.5221183, "chk": "ok", "code_len": 2267}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8730, "tout": 51, "ts": 1791534059.5514593, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8805, "tout": 1067, "ts": 1791534068.408987, "chk": null, "code_len": 2267}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791534028.290543}
{"ep": "h2-00229", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0059, "speedup": 1.0059, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 26.1, "calls": 1, "tokens_in": 6367, "tokens_out": 544, "ctx_chars": 10847, "t_end": 1791534181.4635472, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0059, "lo95": 1.0051, "mean": 1.0059, "ev_only": 1.0065, "ref_ms": 7.94414, "cand_ms": 7.89724, "n_trials": 137, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 16.39, "compile_s": 1.95, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 544, "ts": 1791534164.9760969, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791534155.3626401}
{"ep": "h2-00193", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0049, "speedup": 1.0049, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 128.2, "calls": 2, "tokens_in": 11396, "tokens_out": 1101, "ctx_chars": 9028, "t_end": 1791534185.8760545, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0049, "lo95": 1.0042, "mean": 1.006, "ev_only": 1.003, "ref_ms": 8.07036, "cand_ms": 8.03124, "n_trials": 81, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 49.5, "compile_s": 2.85, "r_error": "", "src_len": 1037, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 546, "ts": 1791534062.8204958, "chk": "ok", "code_len": 1024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6027, "tout": 555, "ts": 1791534136.270261, "chk": null, "code_len": 1037}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791534057.6702492}
{"ep": "h2-00131", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.2935, "speedup": 0.2935, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 318.2, "calls": 4, "tokens_in": 37254, "tokens_out": 8126, "ctx_chars": 6513, "t_end": 1791534190.786426, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.2935, "lo95": 0.2913, "mean": 0.3018, "ev_only": 0.2934, "ref_ms": 7.39916, "cand_ms": 25.21403, "n_trials": 38, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0543, "r_wall_s": 105.69, "compile_s": 3.1, "r_error": "", "src_len": 5795, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2792, "ts": 1791533895.6923444, "chk": "incorrect", "code_len": 6519}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8569, "tout": 2814, "ts": 1791534050.105205, "chk": "ok", "code_len": 5795}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11495, "tout": 26, "ts": 1791534064.6588674, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11545, "tout": 2494, "ts": 1791534084.9931579, "chk": null, "code_len": 5795}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791533872.5912602}
{"ep": "h2-00167", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.6402, "speedup": 2.6402, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 193.9, "calls": 2, "tokens_in": 13037, "tokens_out": 4383, "ctx_chars": 8529, "t_end": 1791534194.1378274, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 2.6402, "lo95": 2.6346, "mean": 2.6352, "ev_only": 2.6552, "ref_ms": 8.51634, "cand_ms": 3.22563, "n_trials": 172, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3414, "r_wall_s": 140.1, "compile_s": 2.57, "r_error": "", "src_len": 2992, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2732, "ts": 1791534022.8893495, "chk": "ok", "code_len": 2992}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7540, "tout": 1651, "ts": 1791534053.9228826, "chk": null, "code_len": 2992}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791534000.2416725}
{"ep": "h2-00221", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 64.2, "calls": 1, "tokens_in": 1501, "tokens_out": 839, "ctx_chars": 0, "t_end": 1791534196.302429, "error": "RuntimeError: Error building extension 'mish2_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mish2_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 55.62, "compile_s": 55.62, "r_error": "RuntimeError: Error building extension 'mish2_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mish2_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "src_len": 1688, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1501, "tout": 839, "ts": 1791534140.6549973, "chk": null, "code_len": 1688}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1501, "started": 1791534132.1211538}
{"ep": "h2-00173", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8196, "speedup": 0.8196, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 189.7, "calls": 3, "tokens_in": 49977, "tokens_out": 14939, "ctx_chars": 2425, "t_end": 1791534200.370654, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.8196, "lo95": 0.7989, "mean": 0.9313, "ev_only": 0.8189, "ref_ms": 19.89957, "cand_ms": 24.27855, "n_trials": 31, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0296, "r_wall_s": 61.03, "compile_s": 2.12, "r_error": "", "src_len": 17582, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7511, "ts": 1791534070.2900352, "chk": "ok", "code_len": 17582}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19171, "tout": 63, "ts": 1791534082.0534801, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19258, "tout": 7365, "ts": 1791534139.313072, "chk": null, "code_len": 17582}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791534010.6607063}
{"ep": "h2-00109", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.784, "speedup": 0.784, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 395.6, "calls": 5, "tokens_in": 75849, "tokens_out": 18117, "ctx_chars": 6513, "t_end": 1791534202.6764076, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.784, "lo95": 0.7727, "mean": 0.789, "ev_only": 0.7813, "ref_ms": 7.1836, "cand_ms": 9.16282, "n_trials": 95, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0948, "r_wall_s": 66.93, "compile_s": 1.97, "r_error": "", "src_len": 5063, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 6285, "ts": 1791533856.6014404, "chk": "incorrect", "code_len": 6822}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 12062, "tout": 4395, "ts": 1791533913.9882755, "chk": "incorrect", "code_len": 6822}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 16589, "tout": 2700, "ts": 1791533950.3116996, "chk": "incorrect", "code_len": 5217}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 19421, "tout": 2579, "ts": 1791534023.1764903, "chk": "ok", "code_len": 5575}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 22132, "tout": 2158, "ts": 1791534135.7185564, "chk": null, "code_len": 5063}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "incorrect", "ok", null], "first_tin": 5645, "started": 1791533807.0752687}
{"ep": "h2-00219", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0618, "speedup": 1.0618, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 83.1, "calls": 2, "tokens_in": 13930, "tokens_out": 3368, "ctx_chars": 10167, "t_end": 1791534205.8458543, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0618, "lo95": 1.0617, "mean": 1.0616, "ev_only": 1.0626, "ref_ms": 8.48634, "cand_ms": 7.99215, "n_trials": 185, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0064, "r_wall_s": 19.65, "compile_s": 2.32, "r_error": "", "src_len": 2445, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 2124, "ts": 1791534140.5147777, "chk": "ok", "code_len": 2445}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7824, "tout": 1244, "ts": 1791534186.127485, "chk": null, "code_len": 2445}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791534122.7265563}
{"ep": "h2-00225", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8435, "speedup": 0.8435, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 63.7, "calls": 2, "tokens_in": 15685, "tokens_out": 2927, "ctx_chars": 11175, "t_end": 1791534209.909908, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8435, "lo95": 0.8425, "mean": 0.8433, "ev_only": 0.8439, "ref_ms": 14.36418, "cand_ms": 17.02886, "n_trials": 72, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 29.97, "compile_s": 1.87, "r_error": "", "src_len": 2374, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1744, "ts": 1791534160.1614447, "chk": "ok", "code_len": 2374}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8653, "tout": 1183, "ts": 1791534179.8538418, "chk": null, "code_len": 2374}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791534146.202647}
{"ep": "h2-00212", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0161, "speedup": 1.0161, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.7, "calls": 4, "tokens_in": 36338, "tokens_out": 4661, "ctx_chars": 11823, "t_end": 1791534210.2017426, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0161, "lo95": 1.0148, "mean": 1.0166, "ev_only": 1.0178, "ref_ms": 27.73725, "cand_ms": 27.29724, "n_trials": 48, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0117, "r_wall_s": 24.76, "compile_s": 2.11, "r_error": "", "src_len": 2499, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1565, "ts": 1791534124.2330778, "chk": "ok", "code_len": 2490}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8598, "tout": 1669, "ts": 1791534163.1592603, "chk": "ok", "code_len": 2499}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10379, "tout": 37, "ts": 1791534173.8456438, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 10440, "tout": 1390, "ts": 1791534185.4178278, "chk": null, "code_len": 2499}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 6921, "started": 1791534109.5204883}
{"ep": "h2-00218", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7205, "speedup": 0.7205, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.7, "calls": 2, "tokens_in": 14295, "tokens_out": 1979, "ctx_chars": 10850, "t_end": 1791534212.6491573, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7205, "lo95": 0.7143, "mean": 0.7177, "ev_only": 0.7264, "ref_ms": 0.15995, "cand_ms": 0.22199, "n_trials": 211, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 39.63, "compile_s": 2.78, "r_error": "", "src_len": 1166, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1271, "ts": 1791534133.9254847, "chk": "ok", "code_len": 1166}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7671, "tout": 708, "ts": 1791534172.6027355, "chk": null, "code_len": 1166}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791534121.9422317}
{"ep": "h2-00142", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 26.7379, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 308.6, "calls": 2, "tokens_in": 7239, "tokens_out": 7663, "ctx_chars": 0, "t_end": 1791534215.299059, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 26.7379, "lo95": 25.837, "mean": 26.422, "ev_only": 27.2231, "ref_ms": 19.67572, "cand_ms": 0.73587, "n_trials": 73, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 176.94, "compile_s": 15.34, "r_error": "", "src_len": 5739, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4968, "ts": 1791533948.666248, "chk": "ok", "code_len": 5739}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5281, "tout": 2695, "ts": 1791534038.233781, "chk": null, "code_len": 5739}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791533906.7206068}
{"ep": "h2-00166", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0184, "speedup": 1.0184, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 218.5, "calls": 3, "tokens_in": 16081, "tokens_out": 2729, "ctx_chars": 5026, "t_end": 1791534216.2366953, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0184, "lo95": 1.0177, "mean": 1.0184, "ev_only": 1.0191, "ref_ms": 8.15244, "cand_ms": 8.00505, "n_trials": 171, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0114, "r_wall_s": 94.09, "compile_s": 1.94, "r_error": "", "src_len": 1551, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1903, "ts": 1791534015.1670258, "chk": "ok", "code_len": 1551}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6008, "tout": 48, "ts": 1791534115.0556803, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6080, "tout": 778, "ts": 1791534122.0510519, "chk": null, "code_len": 1551}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 3993, "started": 1791533997.784397}
{"ep": "h2-00203", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0301, "speedup": 1.0301, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 144.7, "calls": 3, "tokens_in": 27629, "tokens_out": 4770, "ctx_chars": 11065, "t_end": 1791534217.745791, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0301, "lo95": 1.0268, "mean": 1.0285, "ev_only": 1.0327, "ref_ms": 1.83227, "cand_ms": 1.77875, "n_trials": 193, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2736, "r_wall_s": 83.29, "compile_s": 3.18, "r_error": "", "src_len": 2059, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 3737, "ts": 1791534109.1412168, "chk": "ok", "code_len": 2059}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10476, "tout": 26, "ts": 1791534125.7746575, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10526, "tout": 1007, "ts": 1791534134.4139426, "chk": null, "code_len": 2059}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6627, "started": 1791534073.0749063}
{"ep": "h2-00197", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9761, "speedup": 0.9761, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.5, "calls": 1, "tokens_in": 7032, "tokens_out": 998, "ctx_chars": 11539, "t_end": 1791534222.8701599, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9761, "lo95": 0.9756, "mean": 0.9763, "ev_only": 0.9759, "ref_ms": 31.38232, "cand_ms": 32.14958, "n_trials": 19, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0185, "r_wall_s": 143.13, "compile_s": 1.9, "r_error": "", "src_len": 1761, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 998, "ts": 1791534078.600135, "chk": null, "code_len": 1761}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791534068.3272166}
{"ep": "h2-00069", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 7.2053, "speedup": 7.2053, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 516.8, "calls": 3, "tokens_in": 29686, "tokens_out": 24869, "ctx_chars": 8224, "t_end": 1791534225.9315653, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 7.2053, "lo95": 7.1646, "mean": 7.177, "ev_only": 7.1963, "ref_ms": 32.50196, "cand_ms": 4.51085, "n_trials": 42, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9586, "r_wall_s": 46.61, "compile_s": 13.26, "r_error": "", "src_len": 7668, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 16580, "ts": 1791533859.484524, "chk": "no_custom_kernel", "code_len": 7671}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 10021, "tout": 4385, "ts": 1791534059.2292917, "chk": "ok", "code_len": 7668}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 14155, "tout": 3904, "ts": 1791534179.0784101, "chk": null, "code_len": 7668}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791533709.1687136}
{"ep": "h2-00208", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.6423, "speedup": 0.6423, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 139.3, "calls": 2, "tokens_in": 15023, "tokens_out": 10977, "ctx_chars": 6513, "t_end": 1791534228.6075327, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.6423, "lo95": 0.6395, "mean": 0.6369, "ev_only": 0.6417, "ref_ms": 7.36937, "cand_ms": 11.47265, "n_trials": 107, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3483, "r_wall_s": 35.72, "compile_s": 2.06, "r_error": "", "src_len": 6952, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 7645, "ts": 1791534157.1654408, "chk": "ok", "code_len": 7009}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9376, "tout": 3332, "ts": 1791534192.6374285, "chk": null, "code_len": 6952}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791534090.0608058}
{"ep": "h2-00205", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.067, "speedup": 0.067, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.5, "calls": 2, "tokens_in": 14060, "tokens_out": 4763, "ctx_chars": 9970, "t_end": 1791534230.361815, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.067, "lo95": 0.0669, "mean": 0.0672, "ev_only": 0.0669, "ref_ms": 7.82999, "cand_ms": 116.79931, "n_trials": 12, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 89.97, "compile_s": 2.32, "r_error": "", "src_len": 2778, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 3266, "ts": 1791534104.7253544, "chk": "ok", "code_len": 2778}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8060, "tout": 1497, "ts": 1791534139.7091033, "chk": null, "code_len": 2778}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791534075.9025774}
{"ep": "h2-00202", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4294, "speedup": 1.4294, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 160.3, "calls": 2, "tokens_in": 9487, "tokens_out": 9236, "ctx_chars": 0, "t_end": 1791534230.9845183, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.4294, "lo95": 1.4254, "mean": 1.3958, "ev_only": 1.4265, "ref_ms": 3.28658, "cand_ms": 2.29923, "n_trials": 193, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1511, "r_wall_s": 49.5, "compile_s": 13.65, "r_error": "", "src_len": 7266, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 5890, "ts": 1791534119.8297348, "chk": "ok", "code_len": 7266}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6742, "tout": 3346, "ts": 1791534181.2308764, "chk": null, "code_len": 7266}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791534070.6648185}
{"ep": "h2-00211", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.1933, "speedup": 1.1933, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 127.8, "calls": 2, "tokens_in": 5181, "tokens_out": 4292, "ctx_chars": 0, "t_end": 1791534234.7531717, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.1933, "lo95": 1.1929, "mean": 1.1994, "ev_only": 1.1923, "ref_ms": 7.98652, "cand_ms": 6.69295, "n_trials": 171, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 44.09, "compile_s": 2.03, "r_error": "", "src_len": 2744, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 2817, "ts": 1791534131.3645089, "chk": "ok", "code_len": 2744}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3557, "tout": 1475, "ts": 1791534190.5917401, "chk": null, "code_len": 2744}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791534106.9311042}
{"ep": "h2-00181", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0072, "speedup": 1.0072, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 209.3, "calls": 3, "tokens_in": 24688, "tokens_out": 2180, "ctx_chars": 12117, "t_end": 1791534237.0771048, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0072, "lo95": 1.0046, "mean": 1.0072, "ev_only": 1.0072, "ref_ms": 42.48646, "cand_ms": 42.18203, "n_trials": 33, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0, "r_wall_s": 172.42, "compile_s": 1.92, "r_error": "", "src_len": 2289, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1118, "ts": 1791534037.6977098, "chk": "ok", "code_len": 2289}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8627, "tout": 13, "ts": 1791534054.6203601, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8664, "tout": 1049, "ts": 1791534063.7330494, "chk": null, "code_len": 2289}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791534027.8140163}
{"ep": "h2-00201", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0186, "speedup": 1.0186, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 167.5, "calls": 2, "tokens_in": 16256, "tokens_out": 2648, "ctx_chars": 12117, "t_end": 1791534238.1737006, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0186, "lo95": 1.017, "mean": 1.0185, "ev_only": 1.0186, "ref_ms": 42.86701, "cand_ms": 42.08627, "n_trials": 39, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 106.51, "compile_s": 2.85, "r_error": "", "src_len": 2058, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1555, "ts": 1791534083.5640144, "chk": "ok", "code_len": 2058}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8857, "tout": 1093, "ts": 1791534130.9775913, "chk": null, "code_len": 2058}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791534070.641862}
{"ep": "h2-00234", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0069, "speedup": 1.0069, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 72.9, "calls": 3, "tokens_in": 19855, "tokens_out": 2104, "ctx_chars": 9274, "t_end": 1791534239.2227454, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0069, "lo95": 1.0062, "mean": 1.0059, "ev_only": 1.0082, "ref_ms": 8.33517, "cand_ms": 8.27775, "n_trials": 191, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.042, "r_wall_s": 40.91, "compile_s": 1.9, "r_error": "", "src_len": 1926, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1093, "ts": 1791534176.9016118, "chk": "ok", "code_len": 2014}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6995, "tout": 51, "ts": 1791534190.2443871, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7070, "tout": 960, "ts": 1791534198.2439756, "chk": null, "code_len": 1926}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791534166.338844}
{"ep": "h2-00240", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0089, "speedup": 1.0089, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 59.5, "calls": 1, "tokens_in": 1501, "tokens_out": 726, "ctx_chars": 0, "t_end": 1791534239.6604762, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0089, "lo95": 1.0081, "mean": 1.0088, "ev_only": 1.0093, "ref_ms": 8.02845, "cand_ms": 7.95795, "n_trials": 169, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0078, "r_wall_s": 51.63, "compile_s": 2.08, "r_error": "", "src_len": 1229, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1501, "tout": 726, "ts": 1791534187.9405253, "chk": null, "code_len": 1229}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1501, "started": 1791534180.1522174}
{"ep": "h2-00227", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0452, "speedup": 1.0452, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 95.0, "calls": 2, "tokens_in": 5106, "tokens_out": 3173, "ctx_chars": 0, "t_end": 1791534244.0123234, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0452, "lo95": 1.0431, "mean": 1.0463, "ev_only": 1.0454, "ref_ms": 8.42231, "cand_ms": 8.05796, "n_trials": 159, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0046, "r_wall_s": 56.52, "compile_s": 2.08, "r_error": "", "src_len": 1811, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 2136, "ts": 1791534168.0644898, "chk": "ok", "code_len": 1811}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3340, "tout": 1037, "ts": 1791534187.3940444, "chk": null, "code_len": 1811}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791534149.0393193}
{"ep": "h2-00199", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.024, "speedup": 1.024, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 176.0, "calls": 2, "tokens_in": 5824, "tokens_out": 6041, "ctx_chars": 0, "t_end": 1791534246.2656896, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.024, "lo95": 1.0231, "mean": 1.0234, "ev_only": 1.0241, "ref_ms": 30.44235, "cand_ms": 29.72798, "n_trials": 42, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0166, "r_wall_s": 113.44, "compile_s": 1.9, "r_error": "", "src_len": 4048, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 3911, "ts": 1791534103.182848, "chk": "ok", "code_len": 4048}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4199, "tout": 2130, "ts": 1791534132.3726723, "chk": null, "code_len": 4048}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791534070.2194247}
{"ep": "h2-00183", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.7535, "speedup": 4.7535, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 218.4, "calls": 2, "tokens_in": 7336, "tokens_out": 17751, "ctx_chars": 0, "t_end": 1791534247.119609, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 4.7535, "lo95": 4.7445, "mean": 4.994, "ev_only": 4.7535, "ref_ms": 64.34005, "cand_ms": 13.53533, "n_trials": 20, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.4852, "r_wall_s": 42.06, "compile_s": 14.6, "r_error": "", "src_len": 6426, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 14655, "ts": 1791534164.3521495, "chk": "ok", "code_len": 6526}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5590, "tout": 3096, "ts": 1791534204.807674, "chk": null, "code_len": 6426}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791534028.7659442}
{"ep": "h2-00204", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 4.0966, "speedup": 4.0966, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 174.3, "calls": 2, "tokens_in": 8389, "tokens_out": 15073, "ctx_chars": 0, "t_end": 1791534248.3776977, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 4.0966, "lo95": 4.0858, "mean": 4.1188, "ev_only": 4.1554, "ref_ms": 8.64542, "cand_ms": 2.11038, "n_trials": 165, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.999, "r_wall_s": 29.6, "compile_s": 12.43, "r_error": "", "src_len": 8501, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 10879, "ts": 1791534171.370165, "chk": "ok", "code_len": 8501}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6569, "tout": 4194, "ts": 1791534218.7354555, "chk": null, "code_len": 8501}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791534074.0931869}
{"ep": "h2-00170", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0193, "speedup": 1.0193, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 246.0, "calls": 4, "tokens_in": 35695, "tokens_out": 4390, "ctx_chars": 11823, "t_end": 1791534250.1270258, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0193, "lo95": 1.0183, "mean": 1.02, "ev_only": 1.0183, "ref_ms": 27.55763, "cand_ms": 27.03603, "n_trials": 45, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.015, "r_wall_s": 39.31, "compile_s": 2.43, "r_error": "", "src_len": 2669, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1476, "ts": 1791534017.983029, "chk": "ok", "code_len": 2619}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8509, "tout": 1478, "ts": 1791534112.6227262, "chk": "ok", "code_len": 2669}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10099, "tout": 43, "ts": 1791534199.114423, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 10166, "tout": 1393, "ts": 1791534210.5657125, "chk": null, "code_len": 2669}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 6921, "started": 1791534004.1046126}
{"ep": "h2-00249", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0522, "speedup": 1.0522, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 42.0, "calls": 1, "tokens_in": 4862, "tokens_out": 1923, "ctx_chars": 6513, "t_end": 1791534251.9246354, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0522, "lo95": 1.0517, "mean": 1.0522, "ev_only": 1.0529, "ref_ms": 15.71631, "cand_ms": 14.93698, "n_trials": 104, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1725, "r_wall_s": 26.08, "compile_s": 1.99, "r_error": "", "src_len": 3813, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 4862, "tout": 1923, "ts": 1791534225.5290158, "chk": null, "code_len": 3813}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 4862, "started": 1791534209.9101698}
{"ep": "h2-00214", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0039, "speedup": 1.0039, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 140.6, "calls": 2, "tokens_in": 13488, "tokens_out": 1091, "ctx_chars": 10847, "t_end": 1791534255.4818249, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0039, "lo95": 0.9991, "mean": 1.0021, "ev_only": 1.0056, "ref_ms": 8.08736, "cand_ms": 8.05582, "n_trials": 150, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0038, "r_wall_s": 36.04, "compile_s": 1.98, "r_error": "", "src_len": 821, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 641, "ts": 1791534120.7858984, "chk": "ok", "code_len": 821}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7119, "tout": 450, "ts": 1791534219.3461063, "chk": null, "code_len": 821}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791534114.925973}
{"ep": "h2-00223", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0233, "speedup": 1.0233, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 124.7, "calls": 2, "tokens_in": 9117, "tokens_out": 2341, "ctx_chars": 5026, "t_end": 1791534259.255548, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0233, "lo95": 1.0229, "mean": 1.0233, "ev_only": 1.0221, "ref_ms": 8.14926, "cand_ms": 7.9634, "n_trials": 170, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0083, "r_wall_s": 33.77, "compile_s": 2.76, "r_error": "", "src_len": 1230, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1625, "ts": 1791534148.6402128, "chk": "ok", "code_len": 1230}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5122, "tout": 716, "ts": 1791534225.3072398, "chk": null, "code_len": 1230}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791534134.580074}
{"ep": "h2-00108", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4986, "speedup": 1.4986, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 456.4, "calls": 3, "tokens_in": 16476, "tokens_out": 12227, "ctx_chars": 0, "t_end": 1791534259.4375832, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.4986, "lo95": 1.4975, "mean": 1.4989, "ev_only": 1.5009, "ref_ms": 7.77013, "cand_ms": 5.18482, "n_trials": 196, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 121.05, "compile_s": 20.59, "r_error": "", "src_len": 5666, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6099, "ts": 1791533855.978084, "chk": "no_custom_kernel", "code_len": 5706}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5620, "tout": 3258, "ts": 1791534054.572088, "chk": "ok", "code_len": 5666}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8697, "tout": 2870, "ts": 1791534137.2261007, "chk": null, "code_len": 5666}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791533802.99768}
{"ep": "h2-00198", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.3849, "speedup": 0.3849, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 200.0, "calls": 4, "tokens_in": 18481, "tokens_out": 5566, "ctx_chars": 0, "t_end": 1791534269.4775944, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.3849, "lo95": 0.3845, "mean": 0.3851, "ev_only": 0.3848, "ref_ms": 31.40984, "cand_ms": 81.60429, "n_trials": 15, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 75.29, "compile_s": 2.04, "r_error": "", "src_len": 3480, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2076, "tout": 1959, "ts": 1791534087.8310838, "chk": "incorrect", "code_len": 3528}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4168, "tout": 1817, "ts": 1791534157.535305, "chk": "ok", "code_len": 3480}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6094, "tout": 25, "ts": 1791534178.6785047, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6143, "tout": 1765, "ts": 1791534193.266447, "chk": null, "code_len": 3480}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "no_code", null], "first_tin": 2076, "started": 1791534069.5275476}
{"ep": "h2-00226", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.9956, "speedup": 0.9956, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 122.6, "calls": 2, "tokens_in": 4681, "tokens_out": 2324, "ctx_chars": 0, "t_end": 1791534270.726633, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9956, "lo95": 0.9856, "mean": 1.0617, "ev_only": 0.9364, "ref_ms": 0.15343, "cand_ms": 0.15411, "n_trials": 173, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 82.34, "compile_s": 13.28, "r_error": "", "src_len": 1915, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 1274, "ts": 1791534158.8084204, "chk": "ok", "code_len": 1915}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3032, "tout": 1050, "ts": 1791534188.2358756, "chk": null, "code_len": 1915}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791534148.08934}
{"ep": "h2-00187", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.378, "speedup": 1.378, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 234.1, "calls": 2, "tokens_in": 15552, "tokens_out": 12555, "ctx_chars": 6513, "t_end": 1791534273.3941073, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.378, "lo95": 1.3695, "mean": 1.2483, "ev_only": 1.3779, "ref_ms": 7.12973, "cand_ms": 5.17414, "n_trials": 176, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.4433, "r_wall_s": 69.03, "compile_s": 49.09, "r_error": "", "src_len": 8284, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 8827, "ts": 1791534116.5357804, "chk": "ok", "code_len": 8284}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9905, "tout": 3728, "ts": 1791534204.3227255, "chk": null, "code_len": 8284}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791534039.3420298}
{"ep": "h2-00200", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8632, "speedup": 0.8632, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 203.2, "calls": 3, "tokens_in": 49408, "tokens_out": 14181, "ctx_chars": 2425, "t_end": 1791534273.6155922, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8632, "lo95": 0.8268, "mean": 0.8573, "ev_only": 0.8631, "ref_ms": 20.48374, "cand_ms": 23.72916, "n_trials": 62, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0228, "r_wall_s": 72.38, "compile_s": 2.11, "r_error": "", "src_len": 16588, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7238, "ts": 1791534129.5270338, "chk": "ok", "code_len": 16615}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18898, "tout": 40, "ts": 1791534147.2892518, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 18962, "tout": 6903, "ts": 1791534200.8614638, "chk": null, "code_len": 16588}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791534070.3823545}
{"ep": "h2-00189", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.1046, "speedup": 1.1046, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 233.4, "calls": 2, "tokens_in": 13963, "tokens_out": 14321, "ctx_chars": 6513, "t_end": 1791534275.452566, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.1046, "lo95": 1.1032, "mean": 1.1037, "ev_only": 1.1045, "ref_ms": 15.74448, "cand_ms": 14.25404, "n_trials": 78, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.37, "r_wall_s": 86.69, "compile_s": 2.12, "r_error": "", "src_len": 7425, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 10704, "ts": 1791534138.7592354, "chk": "ok", "code_len": 7599}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9099, "tout": 3617, "ts": 1791534188.3720663, "chk": null, "code_len": 7425}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791534042.0841885}
{"ep": "h2-00231", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0568, "speedup": 1.0568, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.5, "calls": 2, "tokens_in": 11753, "tokens_out": 3781, "ctx_chars": 6513, "t_end": 1791534275.7555149, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0568, "lo95": 1.0521, "mean": 1.0604, "ev_only": 1.0552, "ref_ms": 15.85748, "cand_ms": 15.00573, "n_trials": 81, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1441, "r_wall_s": 68.09, "compile_s": 2.72, "r_error": "", "src_len": 3818, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1917, "ts": 1791534182.0739264, "chk": "ok", "code_len": 3818}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6891, "tout": 1864, "ts": 1791534207.475879, "chk": null, "code_len": 3818}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4862, "started": 1791534164.224549}
{"ep": "h2-00210", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.992, "speedup": 0.992, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 170.2, "calls": 3, "tokens_in": 22876, "tokens_out": 3321, "ctx_chars": 10823, "t_end": 1791534276.6160874, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.992, "lo95": 0.9917, "mean": 1.0318, "ev_only": 0.9923, "ref_ms": 30.16856, "cand_ms": 30.41253, "n_trials": 45, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0408, "r_wall_s": 81.61, "compile_s": 2.12, "r_error": "", "src_len": 3221, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1778, "ts": 1791534122.6871498, "chk": "ok", "code_len": 3415}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8243, "tout": 13, "ts": 1791534181.8464215, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8280, "tout": 1530, "ts": 1791534194.5956368, "chk": null, "code_len": 3221}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791534106.4602213}
{"ep": "h2-00215", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0279, "speedup": 1.0279, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 160.8, "calls": 2, "tokens_in": 12732, "tokens_out": 1605, "ctx_chars": 9656, "t_end": 1791534277.4518838, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0279, "lo95": 1.0266, "mean": 1.0281, "ev_only": 1.0267, "ref_ms": 24.96392, "cand_ms": 24.28619, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0289, "r_wall_s": 123.37, "compile_s": 3.92, "r_error": "", "src_len": 1024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 952, "ts": 1791534125.042615, "chk": "ok", "code_len": 1024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6898, "tout": 653, "ts": 1791534153.3139527, "chk": null, "code_len": 1024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791534116.6850698}
{"ep": "h2-00230", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.5543, "speedup": 0.5543, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 122.3, "calls": 2, "tokens_in": 12655, "tokens_out": 3055, "ctx_chars": 9033, "t_end": 1791534279.212331, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.5543, "lo95": 0.5523, "mean": 0.5529, "ev_only": 0.5522, "ref_ms": 15.60822, "cand_ms": 28.15673, "n_trials": 49, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 61.39, "compile_s": 2.91, "r_error": "", "src_len": 2690, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1570, "ts": 1791534172.0054703, "chk": "ok", "code_len": 2690}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7167, "tout": 1485, "ts": 1791534217.2106173, "chk": null, "code_len": 2690}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5488, "started": 1791534156.8690267}
{"ep": "h2-00259", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "forbidden", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 51.6, "calls": 1, "tokens_in": 1964, "tokens_out": 4940, "ctx_chars": 0, "t_end": 1791534282.53761, "error": "static check failed: line 124: getattr needs a literal, non-private attribute name; line 126: getattr needs a literal, non-private attribute name; line 127: get", "f_status": "forbidden", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 0.01, "compile_s": null, "r_error": "static check failed: line 124: getattr needs a literal, non-private attribute name; line 126: getattr needs a literal, non-private attribute name; line 127: get", "src_len": 5076, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1964, "tout": 4940, "ts": 1791534282.485035, "chk": null, "code_len": 5076}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1964, "started": 1791534230.9848123}
{"ep": "h2-00195", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0235, "speedup": 1.0235, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 221.2, "calls": 2, "tokens_in": 12592, "tokens_out": 1693, "ctx_chars": 9656, "t_end": 1791534287.2981582, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0235, "lo95": 1.0221, "mean": 1.024, "ev_only": 1.0264, "ref_ms": 24.95509, "cand_ms": 24.38247, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0227, "r_wall_s": 178.61, "compile_s": 2.76, "r_error": "", "src_len": 1038, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 1075, "ts": 1791534075.7847152, "chk": "ok", "code_len": 1038}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6758, "tout": 618, "ts": 1791534107.9618058, "chk": null, "code_len": 1038}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791534066.1239643}
{"ep": "h2-00224", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7935, "speedup": 0.7935, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 149.0, "calls": 3, "tokens_in": 19095, "tokens_out": 1781, "ctx_chars": 8224, "t_end": 1791534288.0247116, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7935, "lo95": 0.7867, "mean": 0.7354, "ev_only": 0.7932, "ref_ms": 60.80169, "cand_ms": 76.62626, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 51.4, "compile_s": 2.13, "r_error": "", "src_len": 1107, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 1019, "ts": 1791534151.0371594, "chk": "ok", "code_len": 1107}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6656, "tout": 231, "ts": 1791534231.5160232, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6911, "tout": 531, "ts": 1791534236.570019, "chk": null, "code_len": 1107}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791534139.0646791}
{"ep": "h2-00275", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0076, "speedup": 1.0076, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 23.7, "calls": 1, "tokens_in": 6367, "tokens_out": 541, "ctx_chars": 10847, "t_end": 1791534294.4256284, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0076, "lo95": 1.007, "mean": 1.0075, "ev_only": 1.0077, "ref_ms": 7.95676, "cand_ms": 7.89676, "n_trials": 201, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 18.04, "compile_s": 1.86, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 541, "ts": 1791534276.2776184, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791534270.7269251}
{"ep": "h2-00213", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9934, "speedup": 0.9934, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 189.8, "calls": 3, "tokens_in": 22606, "tokens_out": 3079, "ctx_chars": 10823, "t_end": 1791534301.7808504, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9934, "lo95": 0.9929, "mean": 0.9935, "ev_only": 0.9931, "ref_ms": 30.06035, "cand_ms": 30.2587, "n_trials": 27, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0196, "r_wall_s": 71.06, "compile_s": 2.45, "r_error": "", "src_len": 2906, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1624, "ts": 1791534126.4990501, "chk": "ok", "code_len": 2981}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8089, "tout": 51, "ts": 1791534218.4924662, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8164, "tout": 1404, "ts": 1791534230.3046813, "chk": null, "code_len": 2906}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791534111.9395347}
{"ep": "h2-00244", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7174, "speedup": 0.7174, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 115.7, "calls": 2, "tokens_in": 16483, "tokens_out": 4355, "ctx_chars": 12352, "t_end": 1791534309.789981, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.7174, "lo95": 0.7154, "mean": 0.7166, "ev_only": 0.7173, "ref_ms": 64.47454, "cand_ms": 89.87509, "n_trials": 17, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0018, "r_wall_s": 66.25, "compile_s": 1.99, "r_error": "", "src_len": 2145, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 3199, "ts": 1791534223.735908, "chk": "ok", "code_len": 2145}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9153, "tout": 1156, "ts": 1791534243.3940318, "chk": null, "code_len": 2145}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791534194.1380897}
{"ep": "h2-00248", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0041, "speedup": 1.0041, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 104.4, "calls": 2, "tokens_in": 4037, "tokens_out": 2027, "ctx_chars": 0, "t_end": 1791534310.2676067, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0041, "lo95": 1.003, "mean": 1.005, "ev_only": 1.0045, "ref_ms": 7.99898, "cand_ms": 7.96638, "n_trials": 206, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0046, "r_wall_s": 76.74, "compile_s": 2.59, "r_error": "", "src_len": 888, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1444, "ts": 1791534219.306838, "chk": "ok", "code_len": 888}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2534, "tout": 583, "ts": 1791534233.341929, "chk": null, "code_len": 888}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791534205.8461225}
{"ep": "h2-00274", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8305, "speedup": 0.8305, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 52.8, "calls": 1, "tokens_in": 5528, "tokens_out": 829, "ctx_chars": 8224, "t_end": 1791534322.2641578, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8305, "lo95": 0.8264, "mean": 0.8486, "ev_only": 0.8303, "ref_ms": 62.12647, "cand_ms": 74.80886, "n_trials": 19, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 27.17, "compile_s": 1.99, "r_error": "", "src_len": 1117, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5528, "tout": 829, "ts": 1791534294.8294275, "chk": null, "code_len": 1117}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5528, "started": 1791534269.4779203}
{"ep": "h2-00160", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 3.5931, "speedup": 3.5931, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 359.5, "calls": 3, "tokens_in": 16565, "tokens_out": 17554, "ctx_chars": 0, "t_end": 1791534330.925888, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 3.5931, "lo95": 3.5092, "mean": 3.9674, "ev_only": 3.5947, "ref_ms": 33.00226, "cand_ms": 9.18494, "n_trials": 44, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9964, "r_wall_s": 55.76, "compile_s": 12.97, "r_error": "", "src_len": 6407, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 11053, "ts": 1791534075.9102428, "chk": "no_custom_kernel", "code_len": 6431}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5651, "tout": 3384, "ts": 1791534141.4673152, "chk": "ok", "code_len": 6407}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8948, "tout": 3117, "ts": 1791534275.1376097, "chk": null, "code_len": 6407}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791533971.4552267}
{"ep": "h2-00283", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7673, "speedup": 0.7673, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 50.4, "calls": 1, "tokens_in": 6622, "tokens_out": 489, "ctx_chars": 10850, "t_end": 1791534332.924789, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.7673, "lo95": 0.7579, "mean": 0.8092, "ev_only": 0.7658, "ref_ms": 0.15464, "cand_ms": 0.20154, "n_trials": 185, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 44.46, "compile_s": 1.91, "r_error": "", "src_len": 856, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6622, "tout": 489, "ts": 1791534288.3214793, "chk": null, "code_len": 856}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6622, "started": 1791534282.537851}
{"ep": "h2-00255", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3302, "speedup": 1.3302, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 112.2, "calls": 1, "tokens_in": 5998, "tokens_out": 746, "ctx_chars": 9970, "t_end": 1791534335.0335734, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.3302, "lo95": 1.3298, "mean": 1.3303, "ev_only": 1.3322, "ref_ms": 7.80343, "cand_ms": 5.86642, "n_trials": 169, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0904, "r_wall_s": 104.3, "compile_s": 2.59, "r_error": "", "src_len": 1429, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 746, "ts": 1791534229.9262054, "chk": null, "code_len": 1429}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791534222.8704298}
{"ep": "h2-00245", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0198, "speedup": 1.0198, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 142.6, "calls": 2, "tokens_in": 9091, "tokens_out": 2229, "ctx_chars": 5026, "t_end": 1791534338.8581607, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0198, "lo95": 1.0191, "mean": 1.0211, "ev_only": 1.0203, "ref_ms": 8.15734, "cand_ms": 7.99911, "n_trials": 171, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0083, "r_wall_s": 81.3, "compile_s": 2.77, "r_error": "", "src_len": 1304, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1468, "ts": 1791534209.3364718, "chk": "ok", "code_len": 1304}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5096, "tout": 761, "ts": 1791534257.454203, "chk": null, "code_len": 1304}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791534196.3026943}
{"ep": "h2-00246", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.866, "speedup": 0.866, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 144.9, "calls": 2, "tokens_in": 4138, "tokens_out": 1318, "ctx_chars": 0, "t_end": 1791534345.2752159, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.866, "lo95": 0.8445, "mean": 0.8273, "ev_only": 0.7829, "ref_ms": 0.17178, "cand_ms": 0.19837, "n_trials": 189, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 89.73, "compile_s": 3.2, "r_error": "", "src_len": 1055, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 731, "ts": 1791534206.889696, "chk": "ok", "code_len": 1055}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2489, "tout": 587, "ts": 1791534255.3275235, "chk": null, "code_len": 1055}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791534200.3711498}
{"ep": "h2-00271", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9538, "speedup": 0.9538, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.5, "calls": 2, "tokens_in": 8301, "tokens_out": 6483, "ctx_chars": 0, "t_end": 1791534345.935058, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.9538, "lo95": 0.9435, "mean": 0.9149, "ev_only": 0.962, "ref_ms": 3.43865, "cand_ms": 3.60509, "n_trials": 221, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0818, "r_wall_s": 28.52, "compile_s": 1.9, "r_error": "", "src_len": 5288, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 4009, "ts": 1791534288.957613, "chk": "ok", "code_len": 5288}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5556, "tout": 2474, "ts": 1791534317.3891258, "chk": null, "code_len": 5288}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791534255.4821026}
{"ep": "h2-00222", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0139, "speedup": 1.0139, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 215.8, "calls": 3, "tokens_in": 24727, "tokens_out": 2208, "ctx_chars": 12352, "t_end": 1791534348.2666223, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0139, "lo95": 1.0127, "mean": 1.0135, "ev_only": 1.0139, "ref_ms": 64.34746, "cand_ms": 63.46761, "n_trials": 24, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 76.98, "compile_s": 3.09, "r_error": "", "src_len": 1826, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1219, "ts": 1791534144.5781934, "chk": "ok", "code_len": 1826}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8659, "tout": 57, "ts": 1791534263.2862725, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8740, "tout": 932, "ts": 1791534271.1755235, "chk": null, "code_len": 1826}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791534132.4523704}
{"ep": "h2-00243", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.5235, "speedup": 1.5235, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 164.0, "calls": 2, "tokens_in": 6351, "tokens_out": 7806, "ctx_chars": 0, "t_end": 1791534354.7396657, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.5235, "lo95": 1.523, "mean": 1.5214, "ev_only": 1.5249, "ref_ms": 15.47725, "cand_ms": 10.15931, "n_trials": 97, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2719, "r_wall_s": 62.6, "compile_s": 2.59, "r_error": "", "src_len": 4272, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1904, "tout": 5767, "ts": 1791534244.9775715, "chk": "ok", "code_len": 4272}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4447, "tout": 2039, "ts": 1791534291.3923185, "chk": null, "code_len": 4272}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1904, "started": 1791534190.7866776}
{"ep": "h2-00237", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5292, "speedup": 1.5292, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 180.4, "calls": 2, "tokens_in": 9975, "tokens_out": 11357, "ctx_chars": 0, "t_end": 1791534355.6133854, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.5292, "lo95": 1.5204, "mean": 1.6851, "ev_only": 1.5254, "ref_ms": 3.4967, "cand_ms": 2.28659, "n_trials": 173, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1504, "r_wall_s": 35.1, "compile_s": 14.23, "r_error": "", "src_len": 8613, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 7387, "ts": 1791534236.1772354, "chk": "ok", "code_len": 8613}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7230, "tout": 3970, "ts": 1791534320.263227, "chk": null, "code_len": 8613}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791534175.1981165}
{"ep": "h2-00267", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0442, "speedup": 1.0442, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 108.5, "calls": 2, "tokens_in": 5239, "tokens_out": 3300, "ctx_chars": 0, "t_end": 1791534355.6349373, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0442, "lo95": 1.0432, "mean": 1.0435, "ev_only": 1.0442, "ref_ms": 8.36707, "cand_ms": 8.01316, "n_trials": 78, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 22.6, "compile_s": 2.72, "r_error": "", "src_len": 2072, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 2163, "ts": 1791534266.7100937, "chk": "ok", "code_len": 2072}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3473, "tout": 1137, "ts": 1791534332.649087, "chk": null, "code_len": 2072}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791534247.1199093}
{"ep": "h2-00268", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9841, "speedup": 0.9841, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 108.6, "calls": 2, "tokens_in": 15295, "tokens_out": 6809, "ctx_chars": 11065, "t_end": 1791534356.9312658, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9841, "lo95": 0.9815, "mean": 0.9872, "ev_only": 0.9782, "ref_ms": 1.8138, "cand_ms": 1.84304, "n_trials": 216, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 33.9, "compile_s": 1.99, "r_error": "", "src_len": 3315, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 5139, "ts": 1791534294.9560416, "chk": "ok", "code_len": 3315}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8666, "tout": 1670, "ts": 1791534322.780914, "chk": null, "code_len": 3315}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791534248.377975}
{"ep": "h2-00258", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0642, "speedup": 1.0642, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 126.6, "calls": 3, "tokens_in": 16372, "tokens_out": 2272, "ctx_chars": 6513, "t_end": 1791534356.9751103, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0642, "lo95": 1.0639, "mean": 1.0816, "ev_only": 1.0645, "ref_ms": 15.47305, "cand_ms": 14.53935, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1174, "r_wall_s": 68.9, "compile_s": 1.96, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1148, "ts": 1791534240.682606, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5865, "tout": 13, "ts": 1791534278.0097415, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5902, "tout": 1111, "ts": 1791534287.3015857, "chk": null, "code_len": 2422}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4605, "started": 1791534230.3620603}
{"ep": "h2-00286", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0093, "speedup": 1.0093, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 64.2, "calls": 2, "tokens_in": 4116, "tokens_out": 2405, "ctx_chars": 0, "t_end": 1791534358.6057916, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0093, "lo95": 1.0089, "mean": 1.0357, "ev_only": 1.0099, "ref_ms": 8.05998, "cand_ms": 7.98546, "n_trials": 209, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0055, "r_wall_s": 31.72, "compile_s": 2.11, "r_error": "", "src_len": 1166, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1659, "ts": 1791534309.139539, "chk": "ok", "code_len": 1166}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2613, "tout": 746, "ts": 1791534326.7807074, "chk": null, "code_len": 1166}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791534294.4259014}
{"ep": "h2-00233", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.3476, "speedup": 1.3476, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 196.9, "calls": 2, "tokens_in": 6786, "tokens_out": 7268, "ctx_chars": 0, "t_end": 1791534362.555685, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3476, "lo95": 1.3473, "mean": 1.3473, "ev_only": 1.3476, "ref_ms": 31.3985, "cand_ms": 23.29981, "n_trials": 43, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9611, "r_wall_s": 89.4, "compile_s": 2.42, "r_error": "", "src_len": 3731, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 5255, "ts": 1791534211.5210817, "chk": "ok", "code_len": 3731}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4708, "tout": 2013, "ts": 1791534272.3544164, "chk": null, "code_len": 3731}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791534165.6131196}
{"ep": "h2-00263", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8456, "speedup": 0.8456, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 126.7, "calls": 1, "tokens_in": 7030, "tokens_out": 1255, "ctx_chars": 11175, "t_end": 1791534365.888269, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8456, "lo95": 0.8436, "mean": 0.8859, "ev_only": 0.8459, "ref_ms": 14.27762, "cand_ms": 16.88546, "n_trials": 66, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 115.4, "compile_s": 3.71, "r_error": "", "src_len": 2603, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1255, "ts": 1791534250.398129, "chk": null, "code_len": 2603}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791534239.2230136}
{"ep": "h2-00256", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.6832, "speedup": 0.6832, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 147.9, "calls": 2, "tokens_in": 14158, "tokens_out": 1834, "ctx_chars": 10850, "t_end": 1791534373.859916, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.6832, "lo95": 0.6697, "mean": 0.6334, "ev_only": 0.6956, "ref_ms": 0.15443, "cand_ms": 0.22602, "n_trials": 169, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 43.79, "compile_s": 2.8, "r_error": "", "src_len": 1188, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1128, "ts": 1791534236.5784113, "chk": "ok", "code_len": 1188}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7534, "tout": 706, "ts": 1791534329.9181, "chk": null, "code_len": 1188}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791534225.9322102}
{"ep": "h2-00282", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.992, "speedup": 0.992, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 98.4, "calls": 3, "tokens_in": 22455, "tokens_out": 2992, "ctx_chars": 10823, "t_end": 1791534377.6235993, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.992, "lo95": 0.9908, "mean": 0.9921, "ev_only": 0.9916, "ref_ms": 30.11479, "cand_ms": 30.35835, "n_trials": 41, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0384, "r_wall_s": 52.57, "compile_s": 2.54, "r_error": "", "src_len": 2884, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1548, "ts": 1791534294.2868145, "chk": "ok", "code_len": 2884}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8013, "tout": 52, "ts": 1791534312.8906305, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8089, "tout": 1392, "ts": 1791534324.43997, "chk": null, "code_len": 2884}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791534279.2126014}
{"ep": "h2-00236", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.3239, "speedup": 0.3239, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 206.8, "calls": 3, "tokens_in": 22552, "tokens_out": 5899, "ctx_chars": 6513, "t_end": 1791534378.4849794, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.3239, "lo95": 0.3233, "mean": 0.3261, "ev_only": 0.3225, "ref_ms": 3.25267, "cand_ms": 10.04262, "n_trials": 131, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0255, "r_wall_s": 128.28, "compile_s": 3.96, "r_error": "", "src_len": 6060, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2969, "ts": 1791534195.5835378, "chk": "ok", "code_len": 6060}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8527, "tout": 28, "ts": 1791534227.5191853, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8579, "tout": 2902, "ts": 1791534250.1863496, "chk": null, "code_len": 6060}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791534171.6378553}
{"ep": "h2-00209", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 4.9708, "speedup": 4.9708, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 288.5, "calls": 3, "tokens_in": 14375, "tokens_out": 10776, "ctx_chars": 0, "t_end": 1791534378.5760522, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 4.9708, "lo95": 4.968, "mean": 4.9495, "ev_only": 5.0075, "ref_ms": 7.80372, "cand_ms": 1.5699, "n_trials": 180, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 114.29, "compile_s": 18.05, "r_error": "", "src_len": 5280, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 5330, "ts": 1791534136.2398658, "chk": "no_custom_kernel", "code_len": 5302}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4936, "tout": 2805, "ts": 1791534193.97609, "chk": "ok", "code_len": 5280}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7714, "tout": 2641, "ts": 1791534263.5403073, "chk": null, "code_len": 5280}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1725, "started": 1791534090.1036813}
{"ep": "h2-00191", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.3683, "speedup": 4.3683, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 330.0, "calls": 3, "tokens_in": 18104, "tokens_out": 17536, "ctx_chars": 0, "t_end": 1791534380.2792616, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 4.3683, "lo95": 4.361, "mean": 4.3676, "ev_only": 4.3773, "ref_ms": 31.81595, "cand_ms": 7.28332, "n_trials": 39, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9976, "r_wall_s": 112.21, "compile_s": 14.1, "r_error": "", "src_len": 7308, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 10137, "ts": 1791534139.4557927, "chk": "no_custom_kernel", "code_len": 7332}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6202, "tout": 3871, "ts": 1791534208.0986624, "chk": "ok", "code_len": 7308}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9936, "tout": 3528, "ts": 1791534268.0436947, "chk": null, "code_len": 7308}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791534050.3252404}
{"ep": "h2-00280", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.8906, "speedup": 0.8906, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 108.4, "calls": 2, "tokens_in": 5301, "tokens_out": 4457, "ctx_chars": 0, "t_end": 1791534384.9808362, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8906, "lo95": 0.8854, "mean": 0.8939, "ev_only": 0.8915, "ref_ms": 8.0392, "cand_ms": 9.02703, "n_trials": 72, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 55.48, "compile_s": 2.64, "r_error": "", "src_len": 2881, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 2906, "ts": 1791534303.8407373, "chk": "ok", "code_len": 2881}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3677, "tout": 1551, "ts": 1791534329.4237475, "chk": null, "code_len": 2881}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791534276.6163185}
{"ep": "h2-00289", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.3289, "speedup": 1.3289, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 80.2, "calls": 1, "tokens_in": 1723, "tokens_out": 763, "ctx_chars": 0, "t_end": 1791534390.4532447, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.3289, "lo95": 1.3274, "mean": 1.3265, "ev_only": 1.3325, "ref_ms": 7.87003, "cand_ms": 5.9224, "n_trials": 198, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0901, "r_wall_s": 71.68, "compile_s": 2.7, "r_error": "", "src_len": 1479, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1723, "tout": 763, "ts": 1791534318.062217, "chk": null, "code_len": 1479}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1723, "started": 1791534310.2678885}
{"ep": "h2-00217", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.2887, "speedup": 0.2887, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 270.5, "calls": 5, "tokens_in": 54126, "tokens_out": 10682, "ctx_chars": 6513, "t_end": 1791534390.5006256, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.2887, "lo95": 0.2441, "mean": 0.3163, "ev_only": 0.2878, "ref_ms": 8.23201, "cand_ms": 28.51059, "n_trials": 31, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0534, "r_wall_s": 51.24, "compile_s": 2.41, "r_error": "", "src_len": 5574, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2755, "ts": 1791534143.7332659, "chk": "runtime_error", "code_len": 6274}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8493, "tout": 2858, "ts": 1791534279.3061938, "chk": "incorrect", "code_len": 5819}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 11483, "tout": 2620, "ts": 1791534309.6381538, "chk": "ok", "code_len": 5574}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 14215, "tout": 28, "ts": 1791534319.358642, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 14290, "tout": 2421, "ts": 1791534339.2183583, "chk": null, "code_len": 5574}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["runtime_error", "incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791534120.0494113}
{"ep": "h2-00163", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.1173, "speedup": 0.1173, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 399.7, "calls": 5, "tokens_in": 76628, "tokens_out": 17498, "ctx_chars": 8224, "t_end": 1791534392.5526109, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.1173, "lo95": 0.1167, "mean": 0.113, "ev_only": 0.1174, "ref_ms": 32.82707, "cand_ms": 279.75381, "n_trials": 13, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.6406, "r_wall_s": 133.36, "compile_s": 2.38, "r_error": "", "src_len": 4268, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 8774, "ts": 1791534097.7030325, "chk": "forbidden", "code_len": 991}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 14389, "tout": 2126, "ts": 1791534115.1184227, "chk": "forbidden", "code_len": 4184}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 16676, "tout": 2081, "ts": 1791534132.211459, "chk": "incorrect", "code_len": 4112}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 18842, "tout": 2236, "ts": 1791534206.8577232, "chk": "ok", "code_len": 4215}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 21213, "tout": 2281, "ts": 1791534259.0781138, "chk": null, "code_len": 4268}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "forbidden", "incorrect", "ok", null], "first_tin": 5508, "started": 1791533992.8562372}
{"ep": "h2-00241", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.9886, "speedup": 1.9886, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 213.2, "calls": 2, "tokens_in": 15230, "tokens_out": 5774, "ctx_chars": 11065, "t_end": 1791534394.6805577, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.9886, "lo95": 1.9787, "mean": 1.9411, "ev_only": 1.9918, "ref_ms": 1.80038, "cand_ms": 0.90533, "n_trials": 228, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 142.26, "compile_s": 2.05, "r_error": "", "src_len": 2966, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 4334, "ts": 1791534222.0665913, "chk": "ok", "code_len": 2966}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8601, "tout": 1440, "ts": 1791534252.309304, "chk": null, "code_len": 2966}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791534181.4638114}
{"ep": "h2-00287", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7719, "speedup": 0.7719, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 93.7, "calls": 3, "tokens_in": 19004, "tokens_out": 1741, "ctx_chars": 8224, "t_end": 1791534395.4680233, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.7719, "lo95": 0.6172, "mean": 0.793, "ev_only": 0.771, "ref_ms": 80.26749, "cand_ms": 103.98128, "n_trials": 19, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 55.19, "compile_s": 2.66, "r_error": "", "src_len": 1063, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 959, "ts": 1791534313.9422996, "chk": "ok", "code_len": 1063}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6596, "tout": 260, "ts": 1791534335.4430556, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6880, "tout": 522, "ts": 1791534340.2105098, "chk": null, "code_len": 1063}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791534301.7811415}
{"ep": "h2-00284", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0873, "speedup": 1.0873, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 112.9, "calls": 2, "tokens_in": 15448, "tokens_out": 4722, "ctx_chars": 10527, "t_end": 1791534400.2205613, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0873, "lo95": 1.0855, "mean": 1.0918, "ev_only": 1.0895, "ref_ms": 19.77692, "cand_ms": 18.18904, "n_trials": 79, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0437, "r_wall_s": 63.14, "compile_s": 1.98, "r_error": "", "src_len": 3353, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 2987, "ts": 1791534312.6584394, "chk": "ok", "code_len": 3353}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8785, "tout": 1735, "ts": 1791534337.0195956, "chk": null, "code_len": 3353}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791534287.298435}
{"ep": "h2-00269", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4143, "speedup": 1.4143, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 164.3, "calls": 2, "tokens_in": 6638, "tokens_out": 7456, "ctx_chars": 0, "t_end": 1791534414.4171996, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.4143, "lo95": 1.4139, "mean": 1.4145, "ev_only": 1.4141, "ref_ms": 31.66161, "cand_ms": 22.38668, "n_trials": 42, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9997, "r_wall_s": 78.11, "compile_s": 2.79, "r_error": "", "src_len": 3771, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 5423, "ts": 1791534300.8864586, "chk": "ok", "code_len": 3771}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4560, "tout": 2033, "ts": 1791534335.303066, "chk": null, "code_len": 3771}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791534250.1272843}
{"ep": "h2-00179", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.5594, "speedup": 1.5594, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 388.8, "calls": 3, "tokens_in": 32126, "tokens_out": 14602, "ctx_chars": 11539, "t_end": 1791534415.57679, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.5594, "lo95": 1.5589, "mean": 1.5591, "ev_only": 1.5571, "ref_ms": 7.75808, "cand_ms": 4.97515, "n_trials": 222, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 166.37, "compile_s": 12.65, "r_error": "", "src_len": 6199, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 7892, "ts": 1791534096.5891023, "chk": "no_custom_kernel", "code_len": 6219}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 10822, "tout": 3563, "ts": 1791534183.2319314, "chk": "ok", "code_len": 6199}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 14189, "tout": 3147, "ts": 1791534248.4485521, "chk": null, "code_len": 6199}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 7115, "started": 1791534026.7678773}
{"ep": "h2-00279", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.7909, "speedup": 2.7909, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 144.2, "calls": 2, "tokens_in": 14095, "tokens_out": 4775, "ctx_chars": 9970, "t_end": 1791534419.956918, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.7909, "lo95": 2.7883, "mean": 2.781, "ev_only": 2.7881, "ref_ms": 7.8078, "cand_ms": 2.79757, "n_trials": 213, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 75.61, "compile_s": 2.48, "r_error": "", "src_len": 2942, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 3210, "ts": 1791534303.290774, "chk": "ok", "code_len": 2942}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8095, "tout": 1565, "ts": 1791534343.7033527, "chk": null, "code_len": 2942}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791534275.7557528}
{"ep": "h2-00266", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.6271, "speedup": 2.6271, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 175.5, "calls": 2, "tokens_in": 12948, "tokens_out": 4111, "ctx_chars": 8529, "t_end": 1791534421.8052092, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 2.6271, "lo95": 2.624, "mean": 2.6238, "ev_only": 2.6381, "ref_ms": 8.5872, "cand_ms": 3.26865, "n_trials": 169, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.34, "r_wall_s": 14.51, "compile_s": 2.13, "r_error": "", "src_len": 2897, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2536, "ts": 1791534267.2976074, "chk": "ok", "code_len": 2897}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7451, "tout": 1575, "ts": 1791534407.2417536, "chk": null, "code_len": 2897}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791534246.2659912}
{"ep": "h2-00242", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.012, "speedup": 1.012, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 236.3, "calls": 3, "tokens_in": 19793, "tokens_out": 2069, "ctx_chars": 9274, "t_end": 1791534422.143746, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.012, "lo95": 1.0096, "mean": 1.0112, "ev_only": 1.0111, "ref_ms": 8.33943, "cand_ms": 8.24075, "n_trials": 150, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.043, "r_wall_s": 154.7, "compile_s": 3.59, "r_error": "", "src_len": 1937, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1067, "ts": 1791534196.0839496, "chk": "ok", "code_len": 2009}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6969, "tout": 41, "ts": 1791534259.0788941, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7034, "tout": 961, "ts": 1791534267.3672683, "chk": null, "code_len": 1937}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791534185.8763251}
{"ep": "h2-00174", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0032, "speedup": 1.0032, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 413.5, "calls": 3, "tokens_in": 19989, "tokens_out": 2744, "ctx_chars": 9266, "t_end": 1791534426.8908634, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0032, "lo95": 1.0026, "mean": 1.0378, "ev_only": 1.0043, "ref_ms": 30.40564, "cand_ms": 30.30725, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0421, "r_wall_s": 313.79, "compile_s": 2.18, "r_error": "", "src_len": 2609, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1418, "ts": 1791534026.5164056, "chk": "ok", "code_len": 2609}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7144, "tout": 63, "ts": 1791534101.8669593, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7231, "tout": 1263, "ts": 1791534112.598242, "chk": null, "code_len": 2609}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791534013.422259}
{"ep": "h2-00281", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0233, "speedup": 1.0233, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 153.9, "calls": 2, "tokens_in": 12572, "tokens_out": 1442, "ctx_chars": 9656, "t_end": 1791534431.360105, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0233, "lo95": 1.0216, "mean": 1.023, "ev_only": 1.0245, "ref_ms": 25.04102, "cand_ms": 24.46993, "n_trials": 61, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.029, "r_wall_s": 119.34, "compile_s": 2.05, "r_error": "", "src_len": 1038, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 792, "ts": 1791534286.2188263, "chk": "ok", "code_len": 1038}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6738, "tout": 650, "ts": 1791534311.310246, "chk": null, "code_len": 1038}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791534277.4521284}
{"ep": "h2-00228", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.5041, "speedup": 4.5041, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 283.6, "calls": 3, "tokens_in": 13757, "tokens_out": 9524, "ctx_chars": 0, "t_end": 1791534437.1150713, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 4.5041, "lo95": 4.4849, "mean": 4.5952, "ev_only": 4.5308, "ref_ms": 7.9161, "cand_ms": 1.75754, "n_trials": 206, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 118.48, "compile_s": 14.67, "r_error": "", "src_len": 4989, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 4571, "ts": 1791534196.1131208, "chk": "no_custom_kernel", "code_len": 5011}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4736, "tout": 2539, "ts": 1791534277.9931571, "chk": "ok", "code_len": 4989}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7296, "tout": 2414, "ts": 1791534317.9573617, "chk": null, "code_len": 4989}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1725, "started": 1791534154.1860342}
{"ep": "h2-00220", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9951, "speedup": 0.9951, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 314.8, "calls": 4, "tokens_in": 17652, "tokens_out": 2809, "ctx_chars": 0, "t_end": 1791534437.802685, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9951, "lo95": 0.9919, "mean": 0.9943, "ev_only": 0.9944, "ref_ms": 14.32624, "cand_ms": 14.39617, "n_trials": 87, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.0075, "r_wall_s": 182.32, "compile_s": 14.85, "r_error": "", "src_len": 1706, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2181, "tout": 1039, "ts": 1791534134.0536885, "chk": "compile_error", "code_len": 1646}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4480, "tout": 808, "ts": 1791534205.7286172, "chk": "ok", "code_len": 1706}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5400, "tout": 167, "ts": 1791534248.4479225, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 5591, "tout": 795, "ts": 1791534255.305281, "chk": null, "code_len": 1706}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 2181, "started": 1791534122.9661367}
{"ep": "h2-00315", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7907, "speedup": 0.7907, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 43.6, "calls": 2, "tokens_in": 14070, "tokens_out": 1672, "ctx_chars": 10850, "t_end": 1791534438.3214402, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7907, "lo95": 0.7868, "mean": 0.7827, "ev_only": 0.75, "ref_ms": 0.15041, "cand_ms": 0.19021, "n_trials": 208, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 19.24, "compile_s": 3.17, "r_error": "", "src_len": 845, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1142, "ts": 1791534405.2613935, "chk": "ok", "code_len": 845}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7446, "tout": 530, "ts": 1791534418.7096887, "chk": null, "code_len": 845}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791534394.6808121}
{"ep": "h2-00235", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9772, "speedup": 0.9772, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 270.6, "calls": 2, "tokens_in": 12176, "tokens_out": 2878, "ctx_chars": 9033, "t_end": 1791534438.91347, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9772, "lo95": 0.9729, "mean": 0.9766, "ev_only": 0.9784, "ref_ms": 15.54377, "cand_ms": 15.90654, "n_trials": 81, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0163, "r_wall_s": 208.0, "compile_s": 2.31, "r_error": "", "src_len": 1217, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 2098, "ts": 1791534188.2008374, "chk": "ok", "code_len": 1217}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6686, "tout": 780, "ts": 1791534230.6079369, "chk": null, "code_len": 1217}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791534168.3107123}
{"ep": "h2-00316", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0079, "speedup": 1.0079, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 44.9, "calls": 2, "tokens_in": 13720, "tokens_out": 1209, "ctx_chars": 10909, "t_end": 1791534440.3584683, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0079, "lo95": 1.0067, "mean": 1.008, "ev_only": 1.0069, "ref_ms": 8.05817, "cand_ms": 7.9948, "n_trials": 167, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0046, "r_wall_s": 22.14, "compile_s": 2.74, "r_error": "", "src_len": 939, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 653, "ts": 1791534401.299539, "chk": "ok", "code_len": 939}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7241, "tout": 556, "ts": 1791534417.9197857, "chk": null, "code_len": 939}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791534395.4682655}
{"ep": "h2-00247", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.5576, "speedup": 0.5576, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 237.7, "calls": 2, "tokens_in": 12819, "tokens_out": 3105, "ctx_chars": 9033, "t_end": 1791534440.413085, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.5576, "lo95": 0.5574, "mean": 0.5577, "ev_only": 0.5574, "ref_ms": 15.4716, "cand_ms": 27.74559, "n_trials": 49, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 175.36, "compile_s": 3.01, "r_error": "", "src_len": 2505, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1734, "ts": 1791534220.553476, "chk": "ok", "code_len": 2505}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7331, "tout": 1371, "ts": 1791534264.6391137, "chk": null, "code_len": 2505}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5488, "started": 1791534202.6766741}
{"ep": "h2-00314", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0561, "speedup": 1.0561, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 49.1, "calls": 3, "tokens_in": 23154, "tokens_out": 2631, "ctx_chars": 10527, "t_end": 1791534441.624998, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0561, "lo95": 1.0556, "mean": 1.0567, "ev_only": 1.0569, "ref_ms": 19.73737, "cand_ms": 18.68906, "n_trials": 71, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0103, "r_wall_s": 14.13, "compile_s": 2.1, "r_error": "", "src_len": 2372, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1435, "ts": 1791534407.797559, "chk": "ok", "code_len": 2493}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8206, "tout": 57, "ts": 1791534417.8055294, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8287, "tout": 1139, "ts": 1791534427.4395714, "chk": null, "code_len": 2372}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791534392.552996}
{"ep": "h2-00254", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0661, "speedup": 1.0661, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 226.0, "calls": 2, "tokens_in": 4947, "tokens_out": 3701, "ctx_chars": 0, "t_end": 1791534443.7792153, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0661, "lo95": 1.0646, "mean": 1.0647, "ev_only": 1.0635, "ref_ms": 8.51584, "cand_ms": 7.98769, "n_trials": 138, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0053, "r_wall_s": 31.23, "compile_s": 1.97, "r_error": "", "src_len": 2271, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 2435, "ts": 1791534238.689667, "chk": "ok", "code_len": 2271}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3312, "tout": 1266, "ts": 1791534412.252545, "chk": null, "code_len": 2271}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791534217.7460554}
{"ep": "h2-00262", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0267, "speedup": 1.0267, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 206.2, "calls": 1, "tokens_in": 5832, "tokens_out": 538, "ctx_chars": 9656, "t_end": 1791534444.3936174, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0267, "lo95": 1.0259, "mean": 1.0267, "ev_only": 1.0272, "ref_ms": 24.83134, "cand_ms": 24.18529, "n_trials": 49, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0271, "r_wall_s": 200.25, "compile_s": 2.14, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 538, "ts": 1791534243.4426942, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791534238.1740043}
{"ep": "h2-00292", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8143, "speedup": 0.8143, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.9, "calls": 3, "tokens_in": 18871, "tokens_out": 1591, "ctx_chars": 8224, "t_end": 1791534444.8694766, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8143, "lo95": 0.8067, "mean": 0.8437, "ev_only": 0.8142, "ref_ms": 32.97005, "cand_ms": 40.48717, "n_trials": 41, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 27.97, "compile_s": 2.06, "r_error": "", "src_len": 1040, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 1017, "ts": 1791534345.761207, "chk": "ok", "code_len": 1040}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6634, "tout": 71, "ts": 1791534412.212289, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6729, "tout": 503, "ts": 1791534416.859458, "chk": null, "code_len": 1040}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791534332.9250863}
{"ep": "h2-00257", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0047, "speedup": 1.0047, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 222.1, "calls": 2, "tokens_in": 13442, "tokens_out": 1037, "ctx_chars": 10847, "t_end": 1791534450.7093408, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0047, "lo95": 1.0022, "mean": 1.0025, "ev_only": 1.0035, "ref_ms": 7.97828, "cand_ms": 7.94114, "n_trials": 126, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 94.42, "compile_s": 2.0, "r_error": "", "src_len": 821, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 595, "ts": 1791534233.789334, "chk": "ok", "code_len": 821}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7073, "tout": 442, "ts": 1791534356.1742113, "chk": null, "code_len": 821}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791534228.6077993}
{"ep": "h2-00182", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 4.6859, "speedup": 4.6859, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 428.6, "calls": 3, "tokens_in": 28655, "tokens_out": 23531, "ctx_chars": 8224, "t_end": 1791534456.981717, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 4.6859, "lo95": 4.6715, "mean": 4.701, "ev_only": 4.7099, "ref_ms": 33.02933, "cand_ms": 7.04871, "n_trials": 44, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9997, "r_wall_s": 33.03, "compile_s": 12.26, "r_error": "", "src_len": 6733, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 15921, "ts": 1791534174.8995545, "chk": "no_custom_kernel", "code_len": 6736}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9658, "tout": 4068, "ts": 1791534245.3685257, "chk": "ok", "code_len": 6733}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 13487, "tout": 3542, "ts": 1791534423.9046962, "chk": null, "code_len": 6733}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791534028.4017398}
{"ep": "h2-00301", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9193, "speedup": 0.9193, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 107.8, "calls": 2, "tokens_in": 14615, "tokens_out": 9268, "ctx_chars": 6513, "t_end": 1791534464.7672594, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9193, "lo95": 0.8867, "mean": 0.912, "ev_only": 0.9266, "ref_ms": 3.55588, "cand_ms": 3.86787, "n_trials": 195, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0755, "r_wall_s": 22.77, "compile_s": 2.4, "r_error": "", "src_len": 6645, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 6116, "ts": 1791534410.7846286, "chk": "ok", "code_len": 6645}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9167, "tout": 3152, "ts": 1791534441.7452571, "chk": null, "code_len": 6645}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791534356.9315367}
{"ep": "h2-00273", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 218.5, "calls": 1, "tokens_in": 1596, "tokens_out": 2397, "ctx_chars": 0, "t_end": 1791534477.9262, "error": "RuntimeError: Error building extension 'fused_bn039': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_bn039 -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/loca", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 196.01, "compile_s": 196.0, "r_error": "RuntimeError: Error building extension 'fused_bn039': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_bn039 -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/loca", "src_len": 4885, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1596, "tout": 2397, "ts": 1791534281.6649787, "chk": null, "code_len": 4885}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1596, "started": 1791534259.4380534}
{"ep": "h2-00295", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0268, "speedup": 1.0268, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 134.9, "calls": 1, "tokens_in": 5832, "tokens_out": 538, "ctx_chars": 9656, "t_end": 1791534480.1256676, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0268, "lo95": 1.0256, "mean": 1.0267, "ev_only": 1.027, "ref_ms": 24.85208, "cand_ms": 24.20285, "n_trials": 53, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0225, "r_wall_s": 128.37, "compile_s": 3.19, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 538, "ts": 1791534350.3935878, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791534345.2756207}
{"ep": "h2-00307", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0167, "speedup": 1.0167, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 105.1, "calls": 3, "tokens_in": 25156, "tokens_out": 2511, "ctx_chars": 12352, "t_end": 1791534482.7475374, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0167, "lo95": 1.0159, "mean": 1.0431, "ev_only": 1.0168, "ref_ms": 64.50834, "cand_ms": 63.45174, "n_trials": 24, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0024, "r_wall_s": 18.02, "compile_s": 2.07, "r_error": "", "src_len": 2038, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1441, "ts": 1791534391.475624, "chk": "ok", "code_len": 2103}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8881, "tout": 42, "ts": 1791534455.993803, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8947, "tout": 1028, "ts": 1791534464.686832, "chk": null, "code_len": 2038}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791534377.6238644}
{"ep": "h2-00330", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0065, "speedup": 1.0065, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 43.0, "calls": 2, "tokens_in": 4154, "tokens_out": 1828, "ctx_chars": 0, "t_end": 1791534483.4381742, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0065, "lo95": 1.0062, "mean": 1.0057, "ev_only": 1.007, "ref_ms": 7.9395, "cand_ms": 7.88852, "n_trials": 152, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 16.46, "compile_s": 1.85, "r_error": "", "src_len": 960, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 1256, "ts": 1791534452.5616608, "chk": "ok", "code_len": 960}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2530, "tout": 572, "ts": 1791534466.9082742, "chk": null, "code_len": 960}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791534440.413283}
{"ep": "h2-00306", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0898, "speedup": 1.0898, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 115.1, "calls": 2, "tokens_in": 15367, "tokens_out": 4268, "ctx_chars": 10527, "t_end": 1791534488.9511933, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0898, "lo95": 1.0895, "mean": 1.0898, "ev_only": 1.0903, "ref_ms": 19.65891, "cand_ms": 18.03956, "n_trials": 78, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.036, "r_wall_s": 13.58, "compile_s": 1.85, "r_error": "", "src_len": 3137, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 2705, "ts": 1791534398.1640792, "chk": "ok", "code_len": 3137}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8704, "tout": 1563, "ts": 1791534475.2460587, "chk": null, "code_len": 3137}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791534373.860178}
{"ep": "h2-00261", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.9206, "speedup": 1.9206, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.2, "calls": 2, "tokens_in": 30339, "tokens_out": 23216, "ctx_chars": 0, "t_end": 1791534489.296347, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.9206, "lo95": 1.9169, "mean": 1.9343, "ev_only": 1.9227, "ref_ms": 19.02121, "cand_ms": 9.90388, "n_trials": 69, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1661, "r_wall_s": 22.56, "compile_s": 12.18, "r_error": "", "src_len": 19413, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 14636, "ts": 1791534350.6053011, "chk": "ok", "code_len": 19413}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19705, "tout": 8580, "ts": 1791534466.7078078, "chk": null, "code_len": 19413}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791534237.0773888}
{"ep": "h2-00302", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0066, "speedup": 1.0066, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 137.4, "calls": 3, "tokens_in": 20722, "tokens_out": 1005, "ctx_chars": 10909, "t_end": 1791534494.3921556, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0066, "lo95": 1.0059, "mean": 1.0595, "ev_only": 1.0055, "ref_ms": 8.01914, "cand_ms": 7.96641, "n_trials": 147, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 32.8, "compile_s": 2.42, "r_error": "", "src_len": 949, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6477, "tout": 518, "ts": 1791534362.2541966, "chk": "ok", "code_len": 976}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7104, "tout": 13, "ts": 1791534456.7716632, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7141, "tout": 474, "ts": 1791534461.29153, "chk": null, "code_len": 949}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6477, "started": 1791534356.9753816}
{"ep": "h2-00328", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0599, "speedup": 1.0599, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 56.4, "calls": 1, "tokens_in": 1633, "tokens_out": 2695, "ctx_chars": 0, "t_end": 1791534495.3412185, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0599, "lo95": 1.0596, "mean": 1.0594, "ev_only": 1.0608, "ref_ms": 8.49092, "cand_ms": 8.01139, "n_trials": 154, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 25.37, "compile_s": 1.89, "r_error": "", "src_len": 1617, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1633, "tout": 2695, "ts": 1791534469.8906813, "chk": null, "code_len": 1617}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1633, "started": 1791534438.913717}
{"ep": "h2-00251", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.0936, "speedup": 0.0936, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 287.1, "calls": 4, "tokens_in": 26377, "tokens_out": 7688, "ctx_chars": 0, "t_end": 1791534499.777756, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.0936, "lo95": 0.0931, "mean": 0.0939, "ev_only": 0.0936, "ref_ms": 59.04733, "cand_ms": 630.69711, "n_trials": 14, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.2883, "r_wall_s": 26.28, "compile_s": 1.97, "r_error": "", "src_len": 2974, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1984, "tout": 4588, "ts": 1791534255.5243356, "chk": "runtime_error", "code_len": 2928}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6989, "tout": 1551, "ts": 1791534369.076858, "chk": "ok", "code_len": 2974}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8652, "tout": 76, "ts": 1791534461.3498275, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 8752, "tout": 1473, "ts": 1791534473.4436493, "chk": null, "code_len": 2974}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["runtime_error", "ok", "no_code", null], "first_tin": 1984, "started": 1791534212.6494734}
{"ep": "h2-00297", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.3265, "speedup": 0.3265, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 152.3, "calls": 3, "tokens_in": 22095, "tokens_out": 5449, "ctx_chars": 6513, "t_end": 1791534500.560595, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.3265, "lo95": 0.3257, "mean": 0.3283, "ev_only": 0.3252, "ref_ms": 3.23135, "cand_ms": 9.89664, "n_trials": 135, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0251, "r_wall_s": 22.74, "compile_s": 1.88, "r_error": "", "src_len": 5784, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2722, "ts": 1791534370.6146784, "chk": "ok", "code_len": 5784}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8280, "tout": 65, "ts": 1791534457.207783, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8369, "tout": 2662, "ts": 1791534477.7936172, "chk": null, "code_len": 5784}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791534348.2669213}
{"ep": "h2-00265", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0069, "speedup": 1.0069, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 256.8, "calls": 3, "tokens_in": 24613, "tokens_out": 2142, "ctx_chars": 12352, "t_end": 1791534500.7923713, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0069, "lo95": 1.006, "mean": 1.0073, "ev_only": 1.0076, "ref_ms": 64.28031, "cand_ms": 63.84185, "n_trials": 27, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 104.47, "compile_s": 2.85, "r_error": "", "src_len": 1852, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1184, "ts": 1791534255.7707093, "chk": "ok", "code_len": 1852}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8624, "tout": 13, "ts": 1791534388.2913857, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8661, "tout": 945, "ts": 1791534396.280584, "chk": null, "code_len": 1852}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791534244.012617}
{"ep": "h2-00318", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.4429, "speedup": 2.4429, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 88.1, "calls": 3, "tokens_in": 22659, "tokens_out": 4838, "ctx_chars": 8529, "t_end": 1791534502.4954119, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.4429, "lo95": 2.44, "mean": 2.4493, "ev_only": 2.4528, "ref_ms": 8.55958, "cand_ms": 3.50392, "n_trials": 140, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3967, "r_wall_s": 31.29, "compile_s": 1.87, "r_error": "", "src_len": 3623, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2933, "ts": 1791534441.0775034, "chk": "ok", "code_len": 3781}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8540, "tout": 60, "ts": 1791534456.4123201, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8624, "tout": 1845, "ts": 1791534471.1460953, "chk": null, "code_len": 3623}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5495, "started": 1791534414.4174662}
{"ep": "h2-00288", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.788, "speedup": 0.788, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 194.1, "calls": 5, "tokens_in": 44822, "tokens_out": 9904, "ctx_chars": 0, "t_end": 1791534503.917375, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.788, "lo95": 0.7863, "mean": 0.8092, "ev_only": 0.7881, "ref_ms": 3.26963, "cand_ms": 4.14941, "n_trials": 198, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.0149, "r_wall_s": 22.67, "compile_s": 11.18, "r_error": "", "src_len": 3895, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2743, "tout": 3118, "ts": 1791534336.259059, "chk": "compile_error", "code_len": 6644}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 7107, "tout": 2968, "ts": 1791534410.2443824, "chk": "incorrect", "code_len": 6568}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 10206, "tout": 2014, "ts": 1791534447.7608, "chk": "ok", "code_len": 3895}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 12329, "tout": 61, "ts": 1791534466.7953913, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 12437, "tout": 1743, "ts": 1791534481.2121987, "chk": null, "code_len": 3895}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "incorrect", "ok", "no_code", null], "first_tin": 2743, "started": 1791534309.7902517}
{"ep": "h2-00276", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 5.4923, "speedup": 5.4923, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 235.6, "calls": 3, "tokens_in": 18969, "tokens_out": 22055, "ctx_chars": 0, "t_end": 1791534508.969463, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 5.4923, "lo95": 5.4741, "mean": 5.6054, "ev_only": 5.5042, "ref_ms": 32.41071, "cand_ms": 5.90114, "n_trials": 39, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9784, "r_wall_s": 19.75, "compile_s": 11.37, "r_error": "", "src_len": 7660, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 14047, "ts": 1791534400.4912448, "chk": "no_custom_kernel", "code_len": 7663}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6488, "tout": 4225, "ts": 1791534446.3456185, "chk": "ok", "code_len": 7660}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10515, "tout": 3783, "ts": 1791534489.1978583, "chk": null, "code_len": 7660}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791534273.3944106}
{"ep": "h2-00299", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7753, "speedup": 0.7753, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 163.0, "calls": 3, "tokens_in": 50704, "tokens_out": 15562, "ctx_chars": 2425, "t_end": 1791534518.5750763, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.7753, "lo95": 0.7682, "mean": 0.7832, "ev_only": 0.7749, "ref_ms": 19.20127, "cand_ms": 24.76678, "n_trials": 55, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0365, "r_wall_s": 24.11, "compile_s": 1.89, "r_error": "", "src_len": 18120, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7886, "ts": 1791534418.444692, "chk": "ok", "code_len": 18184}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19546, "tout": 40, "ts": 1791534435.2864535, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19610, "tout": 7636, "ts": 1791534494.4016607, "chk": null, "code_len": 18120}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791534355.6136563}
{"ep": "h2-00325", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0045, "speedup": 1.0045, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.7, "calls": 1, "tokens_in": 1622, "tokens_out": 676, "ctx_chars": 0, "t_end": 1791534518.8427045, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0045, "lo95": 1.0038, "mean": 1.0046, "ev_only": 1.0054, "ref_ms": 8.00264, "cand_ms": 7.96702, "n_trials": 79, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0027, "r_wall_s": 74.82, "compile_s": 46.9, "r_error": "", "src_len": 1401, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1622, "tout": 676, "ts": 1791534443.72863, "chk": null, "code_len": 1401}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1622, "started": 1791534437.1153588}
{"ep": "h2-00326", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0055, "speedup": 1.0055, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 88.9, "calls": 1, "tokens_in": 1623, "tokens_out": 2705, "ctx_chars": 0, "t_end": 1791534526.7070904, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0055, "lo95": 1.0049, "mean": 1.0056, "ev_only": 1.0061, "ref_ms": 30.49484, "cand_ms": 30.32895, "n_trials": 57, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0422, "r_wall_s": 64.08, "compile_s": 13.09, "r_error": "", "src_len": 6072, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1623, "tout": 2705, "ts": 1791534461.9784172, "chk": null, "code_len": 6072}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1623, "started": 1791534437.8029528}
{"ep": "h2-00310", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9236, "speedup": 0.9236, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 151.7, "calls": 1, "tokens_in": 2157, "tokens_out": 723, "ctx_chars": 0, "t_end": 1791534531.9341455, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9236, "lo95": 0.923, "mean": 0.9241, "ev_only": 0.9242, "ref_ms": 7.75782, "cand_ms": 8.39975, "n_trials": 180, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0758, "r_wall_s": 142.61, "compile_s": 2.58, "r_error": "", "src_len": 1145, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2157, "tout": 723, "ts": 1791534388.2873147, "chk": null, "code_len": 1145}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2157, "started": 1791534380.279507}
{"ep": "h2-00252", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0082, "speedup": 1.0082, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 318.8, "calls": 2, "tokens_in": 7387, "tokens_out": 10005, "ctx_chars": 0, "t_end": 1791534534.0725048, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0082, "lo95": 1.0074, "mean": 1.0083, "ev_only": 1.0093, "ref_ms": 30.11054, "cand_ms": 29.86592, "n_trials": 41, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0333, "r_wall_s": 155.77, "compile_s": 11.3, "r_error": "", "src_len": 7462, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 6421, "ts": 1791534269.5137994, "chk": "ok", "code_len": 7462}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5789, "tout": 3584, "ts": 1791534377.8552866, "chk": null, "code_len": 7462}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791534215.8353515}
{"ep": "h2-00338", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0029, "speedup": 1.0029, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 63.4, "calls": 2, "tokens_in": 4187, "tokens_out": 1894, "ctx_chars": 0, "t_end": 1791534541.312969, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0029, "lo95": 1.0026, "mean": 1.0035, "ev_only": 1.0015, "ref_ms": 7.93869, "cand_ms": 7.916, "n_trials": 189, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0049, "r_wall_s": 29.48, "compile_s": 1.91, "r_error": "", "src_len": 842, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 1354, "ts": 1791534490.6089435, "chk": "ok", "code_len": 842}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2563, "tout": 540, "ts": 1791534511.7425272, "chk": null, "code_len": 842}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791534477.9264507}
{"ep": "h2-00351", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0044, "speedup": 1.0044, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 36.9, "calls": 1, "tokens_in": 6367, "tokens_out": 531, "ctx_chars": 10847, "t_end": 1791534545.8624146, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0044, "lo95": 1.0028, "mean": 1.0041, "ev_only": 1.0067, "ref_ms": 7.95192, "cand_ms": 7.9167, "n_trials": 188, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 31.29, "compile_s": 2.12, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 531, "ts": 1791534514.464301, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791534508.9697363}
{"ep": "h2-00333", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 10.0, "speedup": 16.0996, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 103.9, "calls": 2, "tokens_in": 15673, "tokens_out": 6378, "ctx_chars": 10527, "t_end": 1791534548.2599776, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 16.0996, "lo95": 16.0377, "mean": 15.9677, "ev_only": 16.3237, "ref_ms": 19.66179, "cand_ms": 1.22126, "n_trials": 62, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "triton", "own_frac": 1.0, "r_wall_s": 38.15, "compile_s": 2.0, "r_error": "", "src_len": 3874, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 4617, "ts": 1791534486.789376, "chk": "ok", "code_len": 3874}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9010, "tout": 1761, "ts": 1791534510.060627, "chk": null, "code_len": 3874}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791534444.3938994}
{"ep": "h2-00293", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0085, "speedup": 1.0085, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 213.9, "calls": 2, "tokens_in": 6295, "tokens_out": 7331, "ctx_chars": 0, "t_end": 1791534548.8949406, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0085, "lo95": 1.0083, "mean": 1.0085, "ev_only": 1.0085, "ref_ms": 29.98712, "cand_ms": 29.73509, "n_trials": 55, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0213, "r_wall_s": 50.8, "compile_s": 2.07, "r_error": "", "src_len": 4955, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 4857, "ts": 1791534377.404844, "chk": "ok", "code_len": 4955}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4697, "tout": 2474, "ts": 1791534497.545956, "chk": null, "code_len": 4955}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791534335.0338418}
{"ep": "h2-00346", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.797, "speedup": 0.797, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 49.5, "calls": 2, "tokens_in": 13845, "tokens_out": 914, "ctx_chars": 10850, "t_end": 1791534549.2691083, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.797, "lo95": 0.7915, "mean": 0.7992, "ev_only": 0.7579, "ref_ms": 0.15205, "cand_ms": 0.19078, "n_trials": 197, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 29.42, "compile_s": 2.55, "r_error": "", "src_len": 808, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 492, "ts": 1791534504.8976653, "chk": "ok", "code_len": 808}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7223, "tout": 422, "ts": 1791534519.7227468, "chk": null, "code_len": 808}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6622, "started": 1791534499.7780304}
{"ep": "h2-00238", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 3.2036, "speedup": 3.2036, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 374.1, "calls": 3, "tokens_in": 27215, "tokens_out": 15152, "ctx_chars": 8224, "t_end": 1791534550.2424495, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 3.2036, "lo95": 2.8636, "mean": 3.412, "ev_only": 3.2051, "ref_ms": 68.37159, "cand_ms": 21.34215, "n_trials": 14, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8731, "r_wall_s": 199.58, "compile_s": 3.17, "r_error": "", "src_len": 6430, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 8544, "ts": 1791534250.1672304, "chk": "no_custom_kernel", "code_len": 6452}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9192, "tout": 3456, "ts": 1791534311.311029, "chk": "ok", "code_len": 6430}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 12493, "tout": 3152, "ts": 1791534350.561703, "chk": null, "code_len": 6430}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791534176.137426}
{"ep": "h2-00323", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4354, "speedup": 1.4354, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 126.0, "calls": 2, "tokens_in": 13110, "tokens_out": 5680, "ctx_chars": 9033, "t_end": 1791534552.8822782, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.4354, "lo95": 1.4341, "mean": 1.4349, "ev_only": 1.4345, "ref_ms": 15.50445, "cand_ms": 10.80111, "n_trials": 105, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9715, "r_wall_s": 56.59, "compile_s": 2.06, "r_error": "", "src_len": 2776, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 3896, "ts": 1791534463.20231, "chk": "ok", "code_len": 2776}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7620, "tout": 1784, "ts": 1791534495.8349848, "chk": null, "code_len": 2776}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791534426.891259}
{"ep": "h2-00329", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 3.7651, "speedup": 3.7651, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 116.8, "calls": 2, "tokens_in": 6139, "tokens_out": 6043, "ctx_chars": 0, "t_end": 1791534557.1450372, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 3.7651, "lo95": 3.6996, "mean": 3.8419, "ev_only": 3.9351, "ref_ms": 1.85159, "cand_ms": 0.49177, "n_trials": 226, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 41.05, "compile_s": 13.01, "r_error": "", "src_len": 4152, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 3899, "ts": 1791534473.3109286, "chk": "ok", "code_len": 4152}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4380, "tout": 2144, "ts": 1791534516.0673995, "chk": null, "code_len": 4152}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1759, "started": 1791534440.3587203}
{"ep": "h2-00332", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0649, "speedup": 1.0649, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 114.8, "calls": 2, "tokens_in": 6258, "tokens_out": 6298, "ctx_chars": 0, "t_end": 1791534558.5904062, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0649, "lo95": 1.0639, "mean": 1.0663, "ev_only": 1.0635, "ref_ms": 8.46408, "cand_ms": 7.94799, "n_trials": 136, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0068, "r_wall_s": 40.77, "compile_s": 14.69, "r_error": "", "src_len": 4811, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 3893, "ts": 1791534475.6278155, "chk": "ok", "code_len": 4846}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4623, "tout": 2405, "ts": 1791534517.725788, "chk": null, "code_len": 4811}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791534443.7794862}
{"ep": "h2-00344", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0389, "speedup": 1.0389, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 68.3, "calls": 3, "tokens_in": 11988, "tokens_out": 4261, "ctx_chars": 0, "t_end": 1791534562.684518, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0389, "lo95": 1.0384, "mean": 1.0388, "ev_only": 1.0402, "ref_ms": 8.51317, "cand_ms": 8.1941, "n_trials": 144, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.022, "r_wall_s": 18.67, "compile_s": 2.19, "r_error": "", "src_len": 1731, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1633, "tout": 3414, "ts": 1791534527.0226917, "chk": "ok", "code_len": 1731}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5159, "tout": 13, "ts": 1791534536.702491, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5196, "tout": 834, "ts": 1791534543.8725145, "chk": null, "code_len": 1731}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1633, "started": 1791534494.3924298}
{"ep": "h2-00353", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.6821, "speedup": 0.6821, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 48.5, "calls": 3, "tokens_in": 21095, "tokens_out": 932, "ctx_chars": 10850, "t_end": 1791534567.2932436, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.6821, "lo95": 0.6628, "mean": 0.6892, "ev_only": 0.6862, "ref_ms": 0.19225, "cand_ms": 0.28187, "n_trials": 217, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 30.6, "compile_s": 1.98, "r_error": "", "src_len": 856, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 487, "ts": 1791534523.8764093, "chk": "ok", "code_len": 856}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7218, "tout": 13, "ts": 1791534532.4140518, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7255, "tout": 432, "ts": 1791534536.5655055, "chk": null, "code_len": 856}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6622, "started": 1791534518.8429873}
{"ep": "h2-00350", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0634, "speedup": 1.0634, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 64.0, "calls": 2, "tokens_in": 13764, "tokens_out": 3123, "ctx_chars": 10167, "t_end": 1791534567.8740919, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0634, "lo95": 1.0629, "mean": 1.0629, "ev_only": 1.0622, "ref_ms": 8.50514, "cand_ms": 7.99771, "n_trials": 164, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0061, "r_wall_s": 30.39, "compile_s": 1.99, "r_error": "", "src_len": 2121, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 1991, "ts": 1791534520.0594614, "chk": "ok", "code_len": 2121}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7658, "tout": 1132, "ts": 1791534537.1733131, "chk": null, "code_len": 2121}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791534503.9176354}
{"ep": "h2-00253", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0113, "speedup": 1.0113, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 353.6, "calls": 2, "tokens_in": 6037, "tokens_out": 7307, "ctx_chars": 0, "t_end": 1791534569.8146503, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0113, "lo95": 1.0106, "mean": 1.0584, "ev_only": 1.0115, "ref_ms": 30.03407, "cand_ms": 29.6971, "n_trials": 35, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0161, "r_wall_s": 183.62, "compile_s": 1.91, "r_error": "", "src_len": 4290, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 5111, "ts": 1791534260.254497, "chk": "ok", "code_len": 4290}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4439, "tout": 2196, "ts": 1791534385.6510422, "chk": null, "code_len": 4290}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791534216.2370374}
{"ep": "h2-00320", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0991, "speedup": 1.0991, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 152.4, "calls": 2, "tokens_in": 16916, "tokens_out": 7246, "ctx_chars": 11539, "t_end": 1791534572.331283, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0991, "lo95": 1.0988, "mean": 1.0989, "ev_only": 1.0995, "ref_ms": 31.63459, "cand_ms": 28.78214, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9684, "r_wall_s": 67.73, "compile_s": 1.88, "r_error": "", "src_len": 4073, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 5044, "ts": 1791534463.01186, "chk": "ok", "code_len": 4073}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9882, "tout": 2202, "ts": 1791534503.7751267, "chk": null, "code_len": 4073}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791534419.9571924}
{"ep": "h2-00303", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4766, "speedup": 1.4766, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 220.7, "calls": 2, "tokens_in": 7805, "tokens_out": 10163, "ctx_chars": 0, "t_end": 1791534579.2902274, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.4766, "lo95": 1.4384, "mean": 1.4522, "ev_only": 1.477, "ref_ms": 15.64538, "cand_ms": 10.59564, "n_trials": 46, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.2788, "r_wall_s": 97.07, "compile_s": 11.03, "r_error": "", "src_len": 7483, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1904, "tout": 6685, "ts": 1791534416.560248, "chk": "ok", "code_len": 7483}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5901, "tout": 3478, "ts": 1791534481.4405255, "chk": null, "code_len": 7483}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1904, "started": 1791534358.606054}
{"ep": "h2-00335", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0672, "speedup": 1.0672, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 137.7, "calls": 2, "tokens_in": 10506, "tokens_out": 2307, "ctx_chars": 6513, "t_end": 1791534588.4209168, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0672, "lo95": 1.0653, "mean": 1.0582, "ev_only": 1.0675, "ref_ms": 15.52351, "cand_ms": 14.54602, "n_trials": 98, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1176, "r_wall_s": 98.67, "compile_s": 1.84, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1184, "ts": 1791534460.969516, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5901, "tout": 1123, "ts": 1791534489.0123131, "chk": null, "code_len": 2422}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791534450.7096107}
{"ep": "h2-00312", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.6513, "speedup": 2.6513, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 209.7, "calls": 3, "tokens_in": 17565, "tokens_out": 16224, "ctx_chars": 0, "t_end": 1791534600.1345193, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 2.6513, "lo95": 2.6481, "mean": 2.6391, "ev_only": 2.6636, "ref_ms": 8.58623, "cand_ms": 3.23848, "n_trials": 179, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9993, "r_wall_s": 49.89, "compile_s": 13.06, "r_error": "", "src_len": 8241, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 8009, "ts": 1791534459.1096249, "chk": "no_custom_kernel", "code_len": 7452}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5816, "tout": 4245, "ts": 1791534507.6563034, "chk": "ok", "code_len": 8241}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9929, "tout": 3970, "ts": 1791534550.1925786, "chk": null, "code_len": 8241}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1820, "started": 1791534390.4536517}
{"ep": "h2-00369", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8699, "speedup": 0.8699, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 32.6, "calls": 1, "tokens_in": 1728, "tokens_out": 2154, "ctx_chars": 0, "t_end": 1791534602.4570634, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8699, "lo95": 0.8684, "mean": 0.8701, "ev_only": 0.8702, "ref_ms": 27.62171, "cand_ms": 31.75184, "n_trials": 41, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1586, "r_wall_s": 12.71, "compile_s": 1.89, "r_error": "", "src_len": 1887, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1728, "tout": 2154, "ts": 1791534589.7244368, "chk": null, "code_len": 1887}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1728, "started": 1791534569.8150811}
{"ep": "h2-00291", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.026, "speedup": 1.026, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 277.7, "calls": 3, "tokens_in": 8655, "tokens_out": 4484, "ctx_chars": 0, "t_end": 1791534608.5899413, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.026, "lo95": 1.0255, "mean": 1.0262, "ev_only": 1.0275, "ref_ms": 8.21777, "cand_ms": 8.0093, "n_trials": 170, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0088, "r_wall_s": 54.45, "compile_s": 2.03, "r_error": "", "src_len": 1357, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 1769, "ts": 1791534347.115582, "chk": "incorrect", "code_len": 1327}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 3001, "tout": 1906, "ts": 1791534538.3194976, "chk": "ok", "code_len": 1357}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 3902, "tout": 809, "ts": 1791534554.0112529, "chk": null, "code_len": 1357}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", null], "first_tin": 1752, "started": 1791534330.9262536}
{"ep": "h2-00277", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.5509, "speedup": 0.5509, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 337.5, "calls": 3, "tokens_in": 20225, "tokens_out": 3279, "ctx_chars": 9033, "t_end": 1791534611.0737152, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.5509, "lo95": 0.5507, "mean": 0.5511, "ev_only": 0.551, "ref_ms": 15.46974, "cand_ms": 28.08028, "n_trials": 50, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 260.31, "compile_s": 3.18, "r_error": "", "src_len": 2751, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1721, "ts": 1791534294.842572, "chk": "ok", "code_len": 2751}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7318, "tout": 77, "ts": 1791534338.2023056, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7419, "tout": 1481, "ts": 1791534350.3916843, "chk": null, "code_len": 2751}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5488, "started": 1791534273.6159778}
{"ep": "h2-00345", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0225, "speedup": 1.0225, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 118.2, "calls": 2, "tokens_in": 6454, "tokens_out": 7388, "ctx_chars": 0, "t_end": 1791534613.5426176, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0225, "lo95": 1.0222, "mean": 1.0225, "ev_only": 1.0224, "ref_ms": 30.50657, "cand_ms": 29.83588, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0168, "r_wall_s": 46.76, "compile_s": 2.2, "r_error": "", "src_len": 5299, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 4659, "ts": 1791534533.3929448, "chk": "ok", "code_len": 5357}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4829, "tout": 2729, "ts": 1791534566.159202, "chk": null, "code_len": 5299}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791534495.3415024}
{"ep": "h2-00294", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 10.0, "speedup": 24.2597, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 280.2, "calls": 2, "tokens_in": 16285, "tokens_out": 8900, "ctx_chars": 10527, "t_end": 1791534619.0951548, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 24.2597, "lo95": 23.1834, "mean": 21.4751, "ev_only": 24.3507, "ref_ms": 19.77101, "cand_ms": 0.81497, "n_trials": 74, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "triton", "own_frac": 1.0, "r_wall_s": 61.91, "compile_s": 2.08, "r_error": "", "src_len": 5293, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 6475, "ts": 1791534397.514724, "chk": "ok", "code_len": 5337}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9622, "tout": 2425, "ts": 1791534556.9276848, "chk": null, "code_len": 5293}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791534339.3818767}
{"ep": "h2-00298", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7031, "speedup": 0.7031, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 265.1, "calls": 2, "tokens_in": 13905, "tokens_out": 1027, "ctx_chars": 10850, "t_end": 1791534619.7992194, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.7031, "lo95": 0.6873, "mean": 0.6926, "ev_only": 0.7678, "ref_ms": 0.14645, "cand_ms": 0.2083, "n_trials": 168, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 196.37, "compile_s": 2.21, "r_error": "", "src_len": 868, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 552, "ts": 1791534361.7128987, "chk": "ok", "code_len": 868}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7283, "tout": 475, "ts": 1791534422.9858103, "chk": null, "code_len": 868}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6622, "started": 1791534354.7402017}
{"ep": "h2-00313", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0267, "speedup": 1.0267, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 229.9, "calls": 2, "tokens_in": 9181, "tokens_out": 2629, "ctx_chars": 5026, "t_end": 1791534620.4057024, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0267, "lo95": 1.0231, "mean": 1.0507, "ev_only": 1.0284, "ref_ms": 8.2652, "cand_ms": 8.05022, "n_trials": 163, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0073, "r_wall_s": 197.28, "compile_s": 2.25, "r_error": "", "src_len": 1418, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1832, "ts": 1791534408.118871, "chk": "ok", "code_len": 1418}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5186, "tout": 797, "ts": 1791534423.005286, "chk": null, "code_len": 1418}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791534390.5008802}
{"ep": "h2-00348", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.3294, "speedup": 0.3294, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 120.8, "calls": 3, "tokens_in": 22515, "tokens_out": 5881, "ctx_chars": 6513, "t_end": 1791534621.5768392, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.3294, "lo95": 0.3272, "mean": 0.3326, "ev_only": 0.3275, "ref_ms": 3.44398, "cand_ms": 10.45437, "n_trials": 128, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0252, "r_wall_s": 41.55, "compile_s": 2.41, "r_error": "", "src_len": 6041, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2940, "ts": 1791534524.050603, "chk": "ok", "code_len": 6041}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8498, "tout": 49, "ts": 1791534556.9539793, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8571, "tout": 2892, "ts": 1791534579.9775374, "chk": null, "code_len": 6041}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791534500.7926128}
{"ep": "h2-00334", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 6.9179, "speedup": 6.9179, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 178.2, "calls": 2, "tokens_in": 6499, "tokens_out": 12764, "ctx_chars": 0, "t_end": 1791534623.0741463, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 6.9179, "lo95": 6.9127, "mean": 6.939, "ev_only": 6.9358, "ref_ms": 64.59543, "cand_ms": 9.33741, "n_trials": 19, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4417, "r_wall_s": 49.28, "compile_s": 2.91, "r_error": "", "src_len": 4456, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 10480, "ts": 1791534544.094531, "chk": "ok", "code_len": 4636}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4753, "tout": 2284, "ts": 1791534573.7491899, "chk": null, "code_len": 4456}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791534444.8697069}
{"ep": "h2-00354", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0137, "speedup": 1.0137, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 101.1, "calls": 3, "tokens_in": 12176, "tokens_out": 5089, "ctx_chars": 0, "t_end": 1791534627.8457432, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0137, "lo95": 1.0131, "mean": 1.0136, "ev_only": 1.014, "ref_ms": 64.37335, "cand_ms": 63.5044, "n_trials": 20, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0022, "r_wall_s": 37.98, "compile_s": 1.9, "r_error": "", "src_len": 3302, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1744, "tout": 3314, "ts": 1791534558.0980244, "chk": "ok", "code_len": 3349}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5170, "tout": 68, "ts": 1791534575.9996238, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5262, "tout": 1707, "ts": 1791534589.7616403, "chk": null, "code_len": 3302}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1744, "started": 1791534527.2545013}
{"ep": "h2-00342", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7172, "speedup": 0.7172, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 141.1, "calls": 2, "tokens_in": 14229, "tokens_out": 1936, "ctx_chars": 10850, "t_end": 1791534630.0177314, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.7172, "lo95": 0.7137, "mean": 0.7107, "ev_only": 0.7185, "ref_ms": 0.15445, "cand_ms": 0.21536, "n_trials": 193, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 54.84, "compile_s": 2.65, "r_error": "", "src_len": 1130, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1242, "ts": 1791534500.3520267, "chk": "ok", "code_len": 1130}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7605, "tout": 694, "ts": 1791534575.0226173, "chk": null, "code_len": 1130}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791534488.9514492}
{"ep": "h2-00207", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.3466, "speedup": 0.3466, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 552.7, "calls": 2, "tokens_in": 16772, "tokens_out": 6529, "ctx_chars": 11539, "t_end": 1791534633.3939037, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.3466, "lo95": 0.3464, "mean": 0.3466, "ev_only": 0.346, "ref_ms": 7.77838, "cand_ms": 22.4433, "n_trials": 58, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.9998, "r_wall_s": 209.92, "compile_s": 2.15, "r_error": "", "src_len": 4002, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 4420, "ts": 1791534117.1390238, "chk": "ok", "code_len": 4002}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9657, "tout": 2109, "ts": 1791534422.6891775, "chk": null, "code_len": 4002}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7115, "started": 1791534080.6677244}
{"ep": "h2-00340", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8219, "speedup": 0.8219, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 151.3, "calls": 3, "tokens_in": 49268, "tokens_out": 14134, "ctx_chars": 2425, "t_end": 1791534634.0071156, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8219, "lo95": 0.82, "mean": 0.8245, "ev_only": 0.8215, "ref_ms": 19.21967, "cand_ms": 23.3847, "n_trials": 60, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0264, "r_wall_s": 26.71, "compile_s": 2.01, "r_error": "", "src_len": 16638, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7164, "ts": 1791534539.6838682, "chk": "ok", "code_len": 16638}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18824, "tout": 48, "ts": 1791534553.424638, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 18896, "tout": 6922, "ts": 1791534607.2692676, "chk": null, "code_len": 16638}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791534482.7478347}
{"ep": "h2-00327", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.8639, "speedup": 1.8639, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 198.5, "calls": 2, "tokens_in": 29128, "tokens_out": 19059, "ctx_chars": 0, "t_end": 1791534636.8041933, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.8639, "lo95": 1.8562, "mean": 1.9288, "ev_only": 1.8688, "ref_ms": 19.28144, "cand_ms": 10.34492, "n_trials": 75, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1344, "r_wall_s": 41.11, "compile_s": 1.88, "r_error": "", "src_len": 16402, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 11759, "ts": 1791534530.0827384, "chk": "ok", "code_len": 16402}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18494, "tout": 7300, "ts": 1791534595.6595807, "chk": null, "code_len": 16402}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791534438.321705}
{"ep": "h2-00352", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9915, "speedup": 0.9915, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 119.1, "calls": 3, "tokens_in": 22373, "tokens_out": 2948, "ctx_chars": 10823, "t_end": 1791534637.655516, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9915, "lo95": 0.9905, "mean": 0.9908, "ev_only": 0.9925, "ref_ms": 30.09793, "cand_ms": 30.35465, "n_trials": 49, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0411, "r_wall_s": 78.79, "compile_s": 2.05, "r_error": "", "src_len": 2870, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1507, "ts": 1791534531.8408682, "chk": "ok", "code_len": 2870}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7972, "tout": 52, "ts": 1791534546.374517, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8048, "tout": 1389, "ts": 1791534557.967575, "chk": null, "code_len": 2870}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791534518.575328}
{"ep": "h2-00364", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0208, "speedup": 1.0208, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 83.4, "calls": 2, "tokens_in": 9814, "tokens_out": 2558, "ctx_chars": 5026, "t_end": 1791534640.5690067, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0208, "lo95": 1.0205, "mean": 1.0204, "ev_only": 1.0196, "ref_ms": 8.18245, "cand_ms": 8.01546, "n_trials": 157, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0089, "r_wall_s": 18.65, "compile_s": 2.23, "r_error": "", "src_len": 1608, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1716, "ts": 1791534573.3273358, "chk": "ok", "code_len": 1608}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5821, "tout": 842, "ts": 1791534621.848276, "chk": null, "code_len": 1608}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791534557.145305}
{"ep": "h2-00264", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0593, "speedup": 1.0593, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 401.6, "calls": 3, "tokens_in": 18677, "tokens_out": 3793, "ctx_chars": 6513, "t_end": 1791534641.3068812, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0593, "lo95": 1.0557, "mean": 1.0556, "ev_only": 1.0551, "ref_ms": 15.86874, "cand_ms": 14.97972, "n_trials": 96, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1483, "r_wall_s": 248.69, "compile_s": 2.9, "r_error": "", "src_len": 3792, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1898, "ts": 1791534255.3302083, "chk": "ok", "code_len": 3792}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6872, "tout": 47, "ts": 1791534377.3450334, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6943, "tout": 1848, "ts": 1791534392.3591466, "chk": null, "code_len": 3792}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791534239.6607213}
{"ep": "h2-00361", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.4607, "speedup": 0.4607, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.4, "calls": 4, "tokens_in": 35394, "tokens_out": 7649, "ctx_chars": 6513, "t_end": 1791534641.6439006, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.4607, "lo95": 0.3657, "mean": 0.3894, "ev_only": 0.4555, "ref_ms": 4.60423, "cand_ms": 9.99336, "n_trials": 140, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0257, "r_wall_s": 21.16, "compile_s": 1.96, "r_error": "", "src_len": 5439, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2477, "ts": 1791534569.5713644, "chk": "forbidden", "code_len": 5165}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8161, "tout": 2585, "ts": 1791534590.6422813, "chk": "ok", "code_len": 5439}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10858, "tout": 47, "ts": 1791534600.639733, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 10929, "tout": 2540, "ts": 1791534620.46423, "chk": null, "code_len": 5439}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", "no_code", null], "first_tin": 5446, "started": 1791534549.2693539}
{"ep": "h2-00370", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9796, "speedup": 0.9796, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 69.7, "calls": 2, "tokens_in": 12159, "tokens_out": 2721, "ctx_chars": 9033, "t_end": 1791534642.033949, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9796, "lo95": 0.9793, "mean": 0.9797, "ev_only": 0.9801, "ref_ms": 15.47409, "cand_ms": 15.79616, "n_trials": 102, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0164, "r_wall_s": 33.08, "compile_s": 2.04, "r_error": "", "src_len": 1179, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 1956, "ts": 1791534591.3894901, "chk": "ok", "code_len": 1179}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6669, "tout": 765, "ts": 1791534608.607471, "chk": null, "code_len": 1179}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791534572.331732}
{"ep": "h2-00359", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0281, "speedup": 1.0281, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 94.7, "calls": 1, "tokens_in": 5832, "tokens_out": 524, "ctx_chars": 9656, "t_end": 1791534642.9811265, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0281, "lo95": 1.0265, "mean": 1.0275, "ev_only": 1.0275, "ref_ms": 24.95934, "cand_ms": 24.27711, "n_trials": 61, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0287, "r_wall_s": 88.68, "compile_s": 2.36, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 524, "ts": 1791534553.3534262, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791534548.2602286}
{"ep": "h2-00317", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3366, "speedup": 1.3366, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 245.2, "calls": 2, "tokens_in": 12816, "tokens_out": 1418, "ctx_chars": 9970, "t_end": 1791534645.4127114, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.3366, "lo95": 1.3347, "mean": 1.3353, "ev_only": 1.3353, "ref_ms": 7.84869, "cand_ms": 5.87228, "n_trials": 206, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0906, "r_wall_s": 215.02, "compile_s": 2.59, "r_error": "", "src_len": 1399, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5998, "tout": 708, "ts": 1791534407.2470996, "chk": "ok", "code_len": 1399}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6818, "tout": 710, "ts": 1791534429.725247, "chk": null, "code_len": 1399}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5998, "started": 1791534400.7589183}
{"ep": "h2-00322", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0056, "speedup": 1.0056, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 226.3, "calls": 3, "tokens_in": 19857, "tokens_out": 2123, "ctx_chars": 9274, "t_end": 1791534648.4742405, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0056, "lo95": 1.0052, "mean": 1.0052, "ev_only": 1.0044, "ref_ms": 8.30589, "cand_ms": 8.25962, "n_trials": 145, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0448, "r_wall_s": 189.97, "compile_s": 1.96, "r_error": "", "src_len": 1974, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1113, "ts": 1791534432.9720583, "chk": "ok", "code_len": 2028}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7015, "tout": 13, "ts": 1791534449.9026213, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7052, "tout": 997, "ts": 1791534458.4290278, "chk": null, "code_len": 1974}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791534422.1440392}
{"ep": "h2-00365", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 3.5253, "speedup": 3.5253, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 91.4, "calls": 2, "tokens_in": 15393, "tokens_out": 5757, "ctx_chars": 11065, "t_end": 1791534650.0236394, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 3.5253, "lo95": 3.5123, "mean": 3.6626, "ev_only": 3.6014, "ref_ms": 1.8119, "cand_ms": 0.51397, "n_trials": 198, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 24.58, "compile_s": 1.89, "r_error": "", "src_len": 3108, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 4096, "ts": 1791534594.6926498, "chk": "ok", "code_len": 3108}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8764, "tout": 1661, "ts": 1791534625.396177, "chk": null, "code_len": 3108}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791534558.5906389}
{"ep": "h2-00270", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.1183, "speedup": 1.1183, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 398.8, "calls": 2, "tokens_in": 12745, "tokens_out": 22134, "ctx_chars": 0, "t_end": 1791534650.7069197, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.1183, "lo95": 1.1143, "mean": 1.1293, "ev_only": 1.1199, "ref_ms": 7.29461, "cand_ms": 6.52293, "n_trials": 213, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9934, "r_wall_s": 65.89, "compile_s": 3.47, "r_error": "", "src_len": 12473, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 15268, "ts": 1791534382.2948706, "chk": "ok", "code_len": 12494}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9801, "tout": 6866, "ts": 1791534584.4421077, "chk": null, "code_len": 12473}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791534251.9249303}
{"ep": "h2-00319", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0049, "speedup": 1.0049, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 235.3, "calls": 2, "tokens_in": 4111, "tokens_out": 2091, "ctx_chars": 0, "t_end": 1791534650.8958783, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0049, "lo95": 1.0036, "mean": 1.0052, "ev_only": 1.0076, "ref_ms": 8.01299, "cand_ms": 7.97422, "n_trials": 89, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 193.74, "compile_s": 1.99, "r_error": "", "src_len": 869, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1543, "ts": 1791534431.1664383, "chk": "ok", "code_len": 869}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2510, "tout": 548, "ts": 1791534457.085438, "chk": null, "code_len": 869}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791534415.5770438}
{"ep": "h2-00290", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0282, "speedup": 1.0282, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 329.1, "calls": 2, "tokens_in": 4505, "tokens_out": 2334, "ctx_chars": 0, "t_end": 1791534651.4043958, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0282, "lo95": 1.0266, "mean": 1.0283, "ev_only": 1.0264, "ref_ms": 24.9205, "cand_ms": 24.23782, "n_trials": 39, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0289, "r_wall_s": 286.09, "compile_s": 2.47, "r_error": "", "src_len": 1124, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1692, "ts": 1791534337.9097881, "chk": "ok", "code_len": 1124}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2837, "tout": 642, "ts": 1791534364.228284, "chk": null, "code_len": 1124}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791534322.264409}
{"ep": "h2-00300", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7855, "speedup": 0.7855, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 297.0, "calls": 3, "tokens_in": 18905, "tokens_out": 1628, "ctx_chars": 8224, "t_end": 1791534652.6395075, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7855, "lo95": 0.7829, "mean": 0.7836, "ev_only": 0.7858, "ref_ms": 59.86218, "cand_ms": 76.20872, "n_trials": 12, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 111.94, "compile_s": 2.95, "r_error": "", "src_len": 951, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 913, "ts": 1791534366.6721053, "chk": "ok", "code_len": 951}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6550, "tout": 253, "ts": 1791534536.1079268, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6827, "tout": 462, "ts": 1791534540.4329236, "chk": null, "code_len": 951}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791534355.6351855}
{"ep": "h2-00358", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0165, "speedup": 1.0165, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.9, "calls": 2, "tokens_in": 16259, "tokens_out": 2605, "ctx_chars": 12117, "t_end": 1791534657.7716906, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0165, "lo95": 1.0139, "mean": 1.0158, "ev_only": 1.0163, "ref_ms": 42.69127, "cand_ms": 41.99994, "n_trials": 29, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 67.13, "compile_s": 1.91, "r_error": "", "src_len": 2040, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1560, "ts": 1791534559.3140178, "chk": "ok", "code_len": 2040}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8860, "tout": 1045, "ts": 1791534589.9112983, "chk": null, "code_len": 2040}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791534545.8626711}
{"ep": "h2-00305", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5282, "speedup": 1.5282, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 294.5, "calls": 2, "tokens_in": 6172, "tokens_out": 7588, "ctx_chars": 0, "t_end": 1791534660.3654943, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.5282, "lo95": 1.5272, "mean": 1.5282, "ev_only": 1.5335, "ref_ms": 8.46192, "cand_ms": 5.53726, "n_trials": 162, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 213.01, "compile_s": 1.97, "r_error": "", "src_len": 4251, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 5289, "ts": 1791534411.4892552, "chk": "ok", "code_len": 4251}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4537, "tout": 2299, "ts": 1791534447.2656078, "chk": null, "code_len": 4251}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791534365.8885446}
{"ep": "h2-00216", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.5712, "speedup": 0.5712, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 541.6, "calls": 2, "tokens_in": 10213, "tokens_out": 19605, "ctx_chars": 0, "t_end": 1791534660.414899, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.5712, "lo95": 0.5696, "mean": 0.5708, "ev_only": 0.5714, "ref_ms": 15.70122, "cand_ms": 27.487, "n_trials": 45, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.1472, "r_wall_s": 237.44, "compile_s": 15.35, "r_error": "", "src_len": 11206, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2161, "tout": 14212, "ts": 1791534239.2292, "chk": "ok", "code_len": 11206}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8052, "tout": 5393, "ts": 1791534422.3991394, "chk": null, "code_len": 11206}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2161, "started": 1791534118.840465}
{"ep": "h2-00232", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9888, "speedup": 0.9888, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 496.4, "calls": 5, "tokens_in": 27604, "tokens_out": 4913, "ctx_chars": 0, "t_end": 1791534661.7637053, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9888, "lo95": 0.9876, "mean": 0.9891, "ev_only": 0.9883, "ref_ms": 42.59201, "cand_ms": 43.07453, "n_trials": 31, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.0169, "r_wall_s": 306.2, "compile_s": 11.7, "r_error": "", "src_len": 2543, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2034, "tout": 1477, "ts": 1791534181.5598567, "chk": "compile_error", "code_len": 2537}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4746, "tout": 1094, "ts": 1791534241.8501937, "chk": "compile_error", "code_len": 2532}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 6054, "tout": 1137, "ts": 1791534309.0006146, "chk": "ok", "code_len": 2543}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7303, "tout": 117, "ts": 1791534345.4376156, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 7467, "tout": 1088, "ts": 1791534354.742467, "chk": null, "code_len": 2543}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "compile_error", "ok", "no_code", null], "first_tin": 2034, "started": 1791534165.4094298}
{"ep": "h2-00372", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0437, "speedup": 1.0437, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 73.4, "calls": 2, "tokens_in": 4968, "tokens_out": 2933, "ctx_chars": 0, "t_end": 1791534661.803781, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0437, "lo95": 1.043, "mean": 1.0635, "ev_only": 1.0441, "ref_ms": 8.28941, "cand_ms": 7.94223, "n_trials": 193, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0062, "r_wall_s": 35.61, "compile_s": 2.44, "r_error": "", "src_len": 1687, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 1963, "ts": 1791534606.623079, "chk": "ok", "code_len": 1687}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3202, "tout": 970, "ts": 1791534626.0152228, "chk": null, "code_len": 1687}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791534588.9523218}
{"ep": "h2-00381", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7525, "speedup": 0.7525, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 44.5, "calls": 1, "tokens_in": 6622, "tokens_out": 478, "ctx_chars": 10850, "t_end": 1791534666.0507789, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.7525, "lo95": 0.7347, "mean": 0.7394, "ev_only": 0.7507, "ref_ms": 0.14671, "cand_ms": 0.19498, "n_trials": 225, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 38.88, "compile_s": 2.48, "r_error": "", "src_len": 827, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6622, "tout": 478, "ts": 1791534626.9781778, "chk": null, "code_len": 827}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6622, "started": 1791534621.577126}
{"ep": "h2-00347", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.1931, "speedup": 3.1931, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 173.9, "calls": 2, "tokens_in": 5402, "tokens_out": 6455, "ctx_chars": 0, "t_end": 1791534674.4654868, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 3.1931, "lo95": 3.1902, "mean": 3.1943, "ev_only": 3.1991, "ref_ms": 27.59103, "cand_ms": 8.64088, "n_trials": 53, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.941, "r_wall_s": 103.65, "compile_s": 2.93, "r_error": "", "src_len": 2852, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 4879, "ts": 1791534542.5142915, "chk": "ok", "code_len": 2852}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3672, "tout": 1576, "ts": 1791534570.7673604, "chk": null, "code_len": 2852}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791534500.5608754}
{"ep": "h2-00371", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.5576, "speedup": 0.5576, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.9, "calls": 2, "tokens_in": 12656, "tokens_out": 3078, "ctx_chars": 9033, "t_end": 1791534680.213751, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.5576, "lo95": 0.5571, "mean": 0.5576, "ev_only": 0.5574, "ref_ms": 15.46978, "cand_ms": 27.74504, "n_trials": 23, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 46.6, "compile_s": 2.02, "r_error": "", "src_len": 2702, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1571, "ts": 1791534593.9407423, "chk": "ok", "code_len": 2702}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7168, "tout": 1507, "ts": 1791534633.1977296, "chk": null, "code_len": 2702}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5488, "started": 1791534579.2905002}
{"ep": "h2-00308", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0655, "speedup": 1.0655, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 302.7, "calls": 2, "tokens_in": 15698, "tokens_out": 4628, "ctx_chars": 10167, "t_end": 1791534681.1682734, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0655, "lo95": 1.0648, "mean": 1.066, "ev_only": 1.0636, "ref_ms": 8.48023, "cand_ms": 7.95928, "n_trials": 150, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0068, "r_wall_s": 249.78, "compile_s": 2.72, "r_error": "", "src_len": 2441, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3378, "ts": 1791534411.5563166, "chk": "ok", "code_len": 2441}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9594, "tout": 1250, "ts": 1791534431.3206482, "chk": null, "code_len": 2441}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6104, "started": 1791534378.4852276}
{"ep": "h2-00296", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0544, "speedup": 1.0544, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 335.3, "calls": 3, "tokens_in": 19620, "tokens_out": 4735, "ctx_chars": 6513, "t_end": 1791534681.1979938, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0544, "lo95": 1.0542, "mean": 1.0543, "ev_only": 1.0534, "ref_ms": 15.70922, "cand_ms": 14.89808, "n_trials": 99, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1725, "r_wall_s": 245.57, "compile_s": 2.46, "r_error": "", "src_len": 4654, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 2362, "ts": 1791534365.2876964, "chk": "ok", "code_len": 4654}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7336, "tout": 62, "ts": 1791534416.9570563, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7422, "tout": 2311, "ts": 1791534435.1117947, "chk": null, "code_len": 4654}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791534345.9353025}
{"ep": "h2-00260", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.1526, "speedup": 1.1526, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 453.7, "calls": 2, "tokens_in": 13046, "tokens_out": 5487, "ctx_chars": 9033, "t_end": 1791534688.4899776, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.1526, "lo95": 1.1514, "mean": 1.1525, "ev_only": 1.1519, "ref_ms": 15.48003, "cand_ms": 13.43013, "n_trials": 94, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9721, "r_wall_s": 316.02, "compile_s": 2.04, "r_error": "", "src_len": 2449, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 3769, "ts": 1791534268.9253042, "chk": "ok", "code_len": 2449}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7556, "tout": 1718, "ts": 1791534372.1316745, "chk": null, "code_len": 2449}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791534234.7534263}
{"ep": "h2-00337", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.6537, "speedup": 3.6537, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 225.1, "calls": 2, "tokens_in": 7717, "tokens_out": 14304, "ctx_chars": 0, "t_end": 1791534689.8289812, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 3.6537, "lo95": 3.6492, "mean": 3.6504, "ev_only": 3.6626, "ref_ms": 8.52596, "cand_ms": 2.33348, "n_trials": 184, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9992, "r_wall_s": 41.29, "compile_s": 12.88, "r_error": "", "src_len": 7158, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 10710, "ts": 1791534560.5359666, "chk": "ok", "code_len": 7158}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5897, "tout": 3594, "ts": 1791534648.5010507, "chk": null, "code_len": 7158}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791534465.2993894}
{"ep": "h2-00311", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8485, "speedup": 0.8485, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 311.4, "calls": 1, "tokens_in": 7030, "tokens_out": 1239, "ctx_chars": 11175, "t_end": 1791534696.4248753, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8485, "lo95": 0.8477, "mean": 0.8492, "ev_only": 0.8476, "ref_ms": 14.3347, "cand_ms": 16.89386, "n_trials": 99, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 300.47, "compile_s": 2.22, "r_error": "", "src_len": 2603, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1239, "ts": 1791534395.839011, "chk": null, "code_len": 2603}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791534384.9811132}
{"ep": "h2-00324", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4086, "speedup": 1.4086, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 272.1, "calls": 3, "tokens_in": 15831, "tokens_out": 13509, "ctx_chars": 0, "t_end": 1791534703.429933, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.4086, "lo95": 1.4078, "mean": 1.4087, "ev_only": 1.4085, "ref_ms": 42.2349, "cand_ms": 29.98262, "n_trials": 34, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9999, "r_wall_s": 89.25, "compile_s": 11.22, "r_error": "", "src_len": 5527, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 7678, "ts": 1791534500.7915425, "chk": "no_custom_kernel", "code_len": 5552}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5451, "tout": 3149, "ts": 1791534558.8111336, "chk": "ok", "code_len": 5527}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8344, "tout": 2682, "ts": 1791534613.5166724, "chk": null, "code_len": 5527}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2036, "started": 1791534431.3603399}
{"ep": "h2-00343", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.1978, "speedup": 1.1978, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 216.4, "calls": 2, "tokens_in": 6497, "tokens_out": 6797, "ctx_chars": 0, "t_end": 1791534705.671654, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.1978, "lo95": 1.1973, "mean": 1.1979, "ev_only": 1.1977, "ref_ms": 31.48446, "cand_ms": 26.28589, "n_trials": 48, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9997, "r_wall_s": 135.1, "compile_s": 2.49, "r_error": "", "src_len": 3320, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 4984, "ts": 1791534533.2442076, "chk": "ok", "code_len": 3320}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4419, "tout": 1813, "ts": 1791534569.6027062, "chk": null, "code_len": 3320}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791534489.2965972}
{"ep": "h2-00336", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 5.0124, "speedup": 5.0124, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.8, "calls": 3, "tokens_in": 15985, "tokens_out": 12655, "ctx_chars": 0, "t_end": 1791534709.8088372, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 5.0124, "lo95": 4.8355, "mean": 5.6553, "ev_only": 5.028, "ref_ms": 94.5665, "cand_ms": 18.86666, "n_trials": 16, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8661, "r_wall_s": 70.18, "compile_s": 12.32, "r_error": "", "src_len": 6125, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 6552, "ts": 1791534514.4512184, "chk": "no_custom_kernel", "code_len": 6176}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5444, "tout": 3162, "ts": 1791534571.4642453, "chk": "ok", "code_len": 6125}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8555, "tout": 2941, "ts": 1791534639.5481725, "chk": null, "code_len": 6125}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791534456.9821846}
{"ep": "h2-00393", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0012, "speedup": 1.0012, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 69.1, "calls": 1, "tokens_in": 6477, "tokens_out": 539, "ctx_chars": 10909, "t_end": 1791534712.0794044, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0012, "lo95": 0.999, "mean": 0.9472, "ev_only": 1.0024, "ref_ms": 8.16109, "cand_ms": 8.15112, "n_trials": 94, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 63.79, "compile_s": 2.17, "r_error": "", "src_len": 1000, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 539, "ts": 1791534648.1242838, "chk": null, "code_len": 1000}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791534642.9813783}
{"ep": "h2-00272", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.6474, "speedup": 0.6474, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 454.2, "calls": 3, "tokens_in": 12998, "tokens_out": 10515, "ctx_chars": 0, "t_end": 1791534713.498742, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.6474, "lo95": 0.6422, "mean": 0.6487, "ev_only": 0.6459, "ref_ms": 7.91546, "cand_ms": 12.22701, "n_trials": 118, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.9993, "r_wall_s": 129.51, "compile_s": 12.74, "r_error": "", "src_len": 4291, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 6040, "ts": 1791534313.6310582, "chk": "no_custom_kernel", "code_len": 4313}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4488, "tout": 2322, "ts": 1791534356.2423048, "chk": "ok", "code_len": 4291}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6785, "tout": 2153, "ts": 1791534583.2858803, "chk": null, "code_len": 4291}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1725, "started": 1791534259.2558117}
{"ep": "h2-00239", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.3923, "speedup": 0.3923, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 534.8, "calls": 4, "tokens_in": 24267, "tokens_out": 8725, "ctx_chars": 0, "t_end": 1791534714.1854718, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.3923, "lo95": 0.3905, "mean": 0.392, "ev_only": 0.3923, "ref_ms": 31.52824, "cand_ms": 80.36214, "n_trials": 17, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 346.74, "compile_s": 2.49, "r_error": "", "src_len": 3439, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2076, "tout": 3227, "ts": 1791534208.49559, "chk": "incorrect", "code_len": 3381}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5436, "tout": 1847, "ts": 1791534297.6896763, "chk": "ok", "code_len": 3399}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 7392, "tout": 1862, "ts": 1791534332.0146844, "chk": "ok", "code_len": 3439}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 9363, "tout": 1789, "ts": 1791534366.6195257, "chk": null, "code_len": 3439}], "n_turns": 4, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "ok", null], "first_tin": 2076, "started": 1791534179.426654}
{"ep": "h2-00418", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0104, "speedup": 1.0104, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 23.9, "calls": 1, "tokens_in": 1599, "tokens_out": 661, "ctx_chars": 0, "t_end": 1791534737.4180899, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0104, "lo95": 1.0101, "mean": 1.0145, "ev_only": 1.0108, "ref_ms": 7.97386, "cand_ms": 7.89193, "n_trials": 168, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0044, "r_wall_s": 17.05, "compile_s": 1.85, "r_error": "", "src_len": 1245, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1599, "tout": 661, "ts": 1791534720.2948723, "chk": null, "code_len": 1245}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1599, "started": 1791534713.499064}
{"ep": "h2-00399", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 87.4, "calls": 1, "tokens_in": 1622, "tokens_out": 821, "ctx_chars": 0, "t_end": 1791534738.7956314, "error": "RuntimeError: Error building extension 'bias_relu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=bias_relu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 79.3, "compile_s": 79.28, "r_error": "RuntimeError: Error building extension 'bias_relu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=bias_relu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "src_len": 1674, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1622, "tout": 821, "ts": 1791534659.4736996, "chk": null, "code_len": 1674}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1622, "started": 1791534651.4046578}
{"ep": "h2-00362", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0569, "speedup": 1.0569, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 200.6, "calls": 2, "tokens_in": 4745, "tokens_out": 3132, "ctx_chars": 0, "t_end": 1791534750.8604612, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0569, "lo95": 1.0554, "mean": 1.0481, "ev_only": 1.0572, "ref_ms": 25.05311, "cand_ms": 23.70471, "n_trials": 54, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0304, "r_wall_s": 100.35, "compile_s": 2.11, "r_error": "", "src_len": 1354, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 2345, "ts": 1791534571.321714, "chk": "ok", "code_len": 1354}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3077, "tout": 787, "ts": 1791534649.686611, "chk": null, "code_len": 1354}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791534550.2426963}
{"ep": "h2-00412", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0595, "speedup": 1.0595, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 70.1, "calls": 2, "tokens_in": 13836, "tokens_out": 3369, "ctx_chars": 10167, "t_end": 1791534759.9568381, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0595, "lo95": 1.059, "mean": 1.0597, "ev_only": 1.0607, "ref_ms": 8.42819, "cand_ms": 7.95481, "n_trials": 133, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0061, "r_wall_s": 17.12, "compile_s": 1.96, "r_error": "", "src_len": 2416, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 2127, "ts": 1791534724.4394588, "chk": "ok", "code_len": 2416}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7730, "tout": 1242, "ts": 1791534742.7449, "chk": null, "code_len": 2416}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791534689.8292332}
{"ep": "h2-00278", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.2157, "speedup": 1.2157, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 485.5, "calls": 2, "tokens_in": 13865, "tokens_out": 13006, "ctx_chars": 6513, "t_end": 1791534761.0007403, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.2157, "lo95": 1.2151, "mean": 1.2388, "ev_only": 1.2171, "ref_ms": 15.73018, "cand_ms": 12.93952, "n_trials": 76, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4808, "r_wall_s": 88.76, "compile_s": 12.66, "r_error": "", "src_len": 7310, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 9402, "ts": 1791534361.3456767, "chk": "ok", "code_len": 7310}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9001, "tout": 3604, "ts": 1791534671.9466145, "chk": null, "code_len": 7310}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791534276.058931}
{"ep": "h2-00304", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.666, "speedup": 2.666, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 398.9, "calls": 4, "tokens_in": 36765, "tokens_out": 6971, "ctx_chars": 8529, "t_end": 1791534761.458331, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.666, "lo95": 2.663, "mean": 2.6581, "ev_only": 2.6765, "ref_ms": 8.53937, "cand_ms": 3.20301, "n_trials": 196, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3362, "r_wall_s": 25.32, "compile_s": 1.98, "r_error": "", "src_len": 3187, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 3616, "ts": 1791534394.06141, "chk": "ok", "code_len": 3641}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9223, "tout": 1657, "ts": 1791534654.6080217, "chk": "ok", "code_len": 3187}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10992, "tout": 39, "ts": 1791534721.2780855, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11055, "tout": 1659, "ts": 1791534736.0865815, "chk": null, "code_len": 3187}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 5495, "started": 1791534362.555959}
{"ep": "h2-00400", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7291, "speedup": 0.7291, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 121.9, "calls": 2, "tokens_in": 14258, "tokens_out": 1968, "ctx_chars": 10850, "t_end": 1791534774.53614, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.7291, "lo95": 0.7236, "mean": 0.7276, "ev_only": 0.7367, "ref_ms": 0.15404, "cand_ms": 0.21127, "n_trials": 204, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 17.55, "compile_s": 1.91, "r_error": "", "src_len": 1145, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1261, "ts": 1791534664.9873106, "chk": "ok", "code_len": 1145}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7634, "tout": 707, "ts": 1791534756.8453445, "chk": null, "code_len": 1145}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791534652.639777}
{"ep": "h2-00411", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9903, "speedup": 0.9903, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 86.4, "calls": 2, "tokens_in": 15668, "tokens_out": 4624, "ctx_chars": 10167, "t_end": 1791534774.8966146, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9903, "lo95": 0.99, "mean": 0.9977, "ev_only": 0.9907, "ref_ms": 8.44544, "cand_ms": 8.52794, "n_trials": 125, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0772, "r_wall_s": 16.1, "compile_s": 1.82, "r_error": "", "src_len": 2448, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3348, "ts": 1791534736.0511587, "chk": "ok", "code_len": 2448}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9564, "tout": 1276, "ts": 1791534758.6401088, "chk": null, "code_len": 2448}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6104, "started": 1791534688.4903908}
{"ep": "h2-00420", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0221, "speedup": 1.0221, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 47.1, "calls": 2, "tokens_in": 9221, "tokens_out": 2502, "ctx_chars": 5026, "t_end": 1791534784.5595438, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0221, "lo95": 1.0215, "mean": 1.0353, "ev_only": 1.0227, "ref_ms": 8.19494, "cand_ms": 8.01804, "n_trials": 152, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0066, "r_wall_s": 16.54, "compile_s": 1.98, "r_error": "", "src_len": 1338, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1726, "ts": 1791534753.1771517, "chk": "ok", "code_len": 1338}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5226, "tout": 776, "ts": 1791534767.9200408, "chk": null, "code_len": 1338}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791534737.4183614}
{"ep": "h2-00391", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0034, "speedup": 1.0034, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 143.8, "calls": 3, "tokens_in": 19846, "tokens_out": 2646, "ctx_chars": 9266, "t_end": 1791534785.4302316, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0034, "lo95": 1.0012, "mean": 1.0321, "ev_only": 1.0034, "ref_ms": 30.4032, "cand_ms": 30.30045, "n_trials": 57, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0383, "r_wall_s": 43.93, "compile_s": 2.17, "r_error": "", "src_len": 2523, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1356, "ts": 1791534653.982461, "chk": "ok", "code_len": 2523}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7082, "tout": 44, "ts": 1791534730.6042793, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7150, "tout": 1246, "ts": 1791534741.0739715, "chk": null, "code_len": 2523}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791534641.6443717}
{"ep": "h2-00423", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8454, "speedup": 0.8454, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 28.3, "calls": 1, "tokens_in": 7030, "tokens_out": 1232, "ctx_chars": 11175, "t_end": 1791534788.2621503, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8454, "lo95": 0.8436, "mean": 0.8455, "ev_only": 0.8456, "ref_ms": 14.29576, "cand_ms": 16.90998, "n_trials": 56, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 17.06, "compile_s": 1.98, "r_error": "", "src_len": 2627, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1232, "ts": 1791534771.113493, "chk": null, "code_len": 2627}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791534759.9571168}
{"ep": "h2-00360", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.4192, "speedup": 0.4192, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 246.7, "calls": 2, "tokens_in": 6326, "tokens_out": 7785, "ctx_chars": 0, "t_end": 1791534795.6029367, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.4192, "lo95": 0.4161, "mean": 0.4215, "ev_only": 0.42, "ref_ms": 7.84566, "cand_ms": 18.71567, "n_trials": 66, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.9995, "r_wall_s": 130.77, "compile_s": 26.1, "r_error": "", "src_len": 4613, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 5608, "ts": 1791534599.3841858, "chk": "ok", "code_len": 4613}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4601, "tout": 2177, "ts": 1791534663.8851976, "chk": null, "code_len": 4613}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1725, "started": 1791534548.895177}
{"ep": "h2-00376", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.4986, "speedup": 0.4986, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 187.8, "calls": 2, "tokens_in": 14419, "tokens_out": 2268, "ctx_chars": 10850, "t_end": 1791534798.8907473, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.4986, "lo95": 0.4956, "mean": 0.5599, "ev_only": 0.5587, "ref_ms": 0.14832, "cand_ms": 0.29749, "n_trials": 183, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 148.03, "compile_s": 4.37, "r_error": "", "src_len": 1403, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1453, "ts": 1791534625.1327178, "chk": "ok", "code_len": 1403}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7795, "tout": 815, "ts": 1791534650.734862, "chk": null, "code_len": 1403}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791534611.0740068}
{"ep": "h2-00363", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7691, "speedup": 0.7691, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 255.1, "calls": 5, "tokens_in": 65748, "tokens_out": 13094, "ctx_chars": 6513, "t_end": 1791534807.978145, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.7691, "lo95": 0.767, "mean": 0.7677, "ev_only": 0.7687, "ref_ms": 7.11283, "cand_ms": 9.24872, "n_trials": 148, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0057, "r_wall_s": 13.93, "compile_s": 2.44, "r_error": "", "src_len": 4692, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 5355, "ts": 1791534595.5682254, "chk": "incorrect", "code_len": 6223}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 11132, "tout": 3449, "ts": 1791534641.2570648, "chk": "incorrect", "code_len": 5276}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 14713, "tout": 2246, "ts": 1791534688.2465162, "chk": "ok", "code_len": 4692}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 17068, "tout": 75, "ts": 1791534777.6825035, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 17190, "tout": 1969, "ts": 1791534794.02065, "chk": null, "code_len": 4692}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791534552.8825607}
{"ep": "h2-00378", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 10.0, "speedup": 14.8663, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 190.9, "calls": 2, "tokens_in": 15448, "tokens_out": 5789, "ctx_chars": 10527, "t_end": 1791534809.98801, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 14.8663, "lo95": 14.8281, "mean": 14.7707, "ev_only": 14.859, "ref_ms": 19.62352, "cand_ms": 1.32, "n_trials": 69, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "triton", "own_frac": 1.0, "r_wall_s": 21.86, "compile_s": 2.07, "r_error": "", "src_len": 3487, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 4239, "ts": 1791534657.561004, "chk": "ok", "code_len": 3487}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8785, "tout": 1550, "ts": 1791534788.0925267, "chk": null, "code_len": 3487}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791534619.0954208}
{"ep": "h2-00413", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9942, "speedup": 0.9942, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 116.8, "calls": 3, "tokens_in": 8315, "tokens_out": 2995, "ctx_chars": 0, "t_end": 1791534813.271484, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9942, "lo95": 0.9934, "mean": 0.9945, "ev_only": 0.9946, "ref_ms": 30.12753, "cand_ms": 30.30199, "n_trials": 45, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0378, "r_wall_s": 72.23, "compile_s": 1.89, "r_error": "", "src_len": 2855, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1596, "tout": 1633, "ts": 1791534711.6074631, "chk": "ok", "code_len": 2855}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3341, "tout": 13, "ts": 1791534727.301324, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 3378, "tout": 1349, "ts": 1791534740.3042812, "chk": null, "code_len": 2855}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1596, "started": 1791534696.425163}
{"ep": "h2-00419", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.074, "speedup": 0.074, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.8, "calls": 4, "tokens_in": 33825, "tokens_out": 5791, "ctx_chars": 8224, "t_end": 1791534814.9590702, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.074, "lo95": 0.0727, "mean": 0.0743, "ev_only": 0.074, "ref_ms": 61.54624, "cand_ms": 832.07803, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1685, "r_wall_s": 29.41, "compile_s": 1.95, "r_error": "", "src_len": 3106, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 2428, "ts": 1791534741.7474682, "chk": "forbidden", "code_len": 3648}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8201, "tout": 1672, "ts": 1791534756.719135, "chk": "ok", "code_len": 3183}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9985, "tout": 102, "ts": 1791534772.2331128, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 10111, "tout": 1589, "ts": 1791534785.4870865, "chk": null, "code_len": 3106}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", "no_code", null], "first_tin": 5528, "started": 1791534714.185932}
{"ep": "h2-00417", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0636, "speedup": 1.0636, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 104.1, "calls": 3, "tokens_in": 16433, "tokens_out": 2321, "ctx_chars": 6513, "t_end": 1791534816.1395943, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0636, "lo95": 1.0629, "mean": 1.0622, "ev_only": 1.0643, "ref_ms": 15.48593, "cand_ms": 14.56009, "n_trials": 95, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1176, "r_wall_s": 62.88, "compile_s": 2.05, "r_error": "", "src_len": 2454, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1172, "ts": 1791534723.7222166, "chk": "ok", "code_len": 2454}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5889, "tout": 26, "ts": 1791534742.6559155, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5939, "tout": 1123, "ts": 1791534752.268564, "chk": null, "code_len": 2454}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4605, "started": 1791534712.0796618}
{"ep": "h2-00375", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4239, "speedup": 1.4239, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 211.0, "calls": 2, "tokens_in": 29141, "tokens_out": 18930, "ctx_chars": 0, "t_end": 1791534819.5958073, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.4239, "lo95": 1.4092, "mean": 1.4072, "ev_only": 1.4267, "ref_ms": 18.4226, "cand_ms": 12.93784, "n_trials": 69, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1338, "r_wall_s": 33.32, "compile_s": 3.0, "r_error": "", "src_len": 16366, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 11546, "ts": 1791534694.917497, "chk": "ok", "code_len": 16366}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18507, "tout": 7384, "ts": 1791534786.1172035, "chk": null, "code_len": 16366}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791534608.5902438}
{"ep": "h2-00422", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0061, "speedup": 1.0061, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 69.6, "calls": 3, "tokens_in": 19800, "tokens_out": 2079, "ctx_chars": 9274, "t_end": 1791534820.499634, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0061, "lo95": 1.0058, "mean": 1.011, "ev_only": 1.0064, "ref_ms": 8.28671, "cand_ms": 8.23674, "n_trials": 141, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.043, "r_wall_s": 41.4, "compile_s": 1.88, "r_error": "", "src_len": 1904, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1053, "ts": 1791534760.7292979, "chk": "ok", "code_len": 1993}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6955, "tout": 76, "ts": 1791534770.9777448, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7055, "tout": 950, "ts": 1791534779.0045922, "chk": null, "code_len": 1904}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791534750.8607235}
{"ep": "h2-00341", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.8422, "speedup": 3.8422, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 349.2, "calls": 3, "tokens_in": 16283, "tokens_out": 15400, "ctx_chars": 0, "t_end": 1791534832.6422548, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 3.8422, "lo95": 3.8272, "mean": 4.5814, "ev_only": 3.8513, "ref_ms": 60.78048, "cand_ms": 15.81908, "n_trials": 23, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8439, "r_wall_s": 41.26, "compile_s": 11.17, "r_error": "", "src_len": 6083, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 9116, "ts": 1791534563.4169385, "chk": "no_custom_kernel", "code_len": 6105}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5535, "tout": 3261, "ts": 1791534651.649237, "chk": "ok", "code_len": 6083}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8762, "tout": 3023, "ts": 1791534791.28153, "chk": null, "code_len": 6083}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791534483.4385242}
{"ep": "h2-00421", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.006, "speedup": 1.006, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 102.8, "calls": 3, "tokens_in": 20077, "tokens_out": 2807, "ctx_chars": 9266, "t_end": 1791534841.603665, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.006, "lo95": 1.0058, "mean": 1.0061, "ev_only": 1.0062, "ref_ms": 30.3729, "cand_ms": 30.1904, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0389, "r_wall_s": 65.22, "compile_s": 1.82, "r_error": "", "src_len": 2654, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1469, "ts": 1791534752.056863, "chk": "ok", "code_len": 2654}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7195, "tout": 49, "ts": 1791534764.7774892, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7268, "tout": 1289, "ts": 1791534775.7321308, "chk": null, "code_len": 2654}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791534738.7959075}
{"ep": "h2-00321", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8444, "speedup": 0.8444, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 421.2, "calls": 2, "tokens_in": 15634, "tokens_out": 2927, "ctx_chars": 11175, "t_end": 1791534842.996521, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8444, "lo95": 0.8435, "mean": 0.8445, "ev_only": 0.8456, "ref_ms": 14.3683, "cand_ms": 17.01639, "n_trials": 75, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 166.24, "compile_s": 10.99, "r_error": "", "src_len": 2472, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1673, "ts": 1791534435.0678244, "chk": "ok", "code_len": 2472}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8602, "tout": 1254, "ts": 1791534676.6115127, "chk": null, "code_len": 2472}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791534421.8054638}
{"ep": "h2-00395", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.9913, "speedup": 0.9913, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 196.6, "calls": 2, "tokens_in": 4191, "tokens_out": 1972, "ctx_chars": 0, "t_end": 1791534845.0316098, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9913, "lo95": 0.9883, "mean": 0.9959, "ev_only": 0.9945, "ref_ms": 8.0039, "cand_ms": 8.07424, "n_trials": 125, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 58.66, "compile_s": 2.91, "r_error": "", "src_len": 864, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 1452, "ts": 1791534663.3476593, "chk": "ok", "code_len": 864}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2567, "tout": 520, "ts": 1791534786.2101893, "chk": null, "code_len": 864}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791534649.0847402}
{"ep": "h2-00390", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9224, "speedup": 0.9224, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 208.2, "calls": 2, "tokens_in": 15315, "tokens_out": 1779, "ctx_chars": 11539, "t_end": 1791534849.480255, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9224, "lo95": 0.9219, "mean": 0.9227, "ev_only": 0.9215, "ref_ms": 7.76677, "cand_ms": 8.41998, "n_trials": 138, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0608, "r_wall_s": 90.93, "compile_s": 2.16, "r_error": "", "src_len": 1736, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 980, "ts": 1791534650.858886, "chk": "ok", "code_len": 1736}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8202, "tout": 799, "ts": 1791534757.6698122, "chk": null, "code_len": 1736}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7113, "started": 1791534641.3071277}
{"ep": "h2-00405", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0201, "speedup": 1.0201, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 188.2, "calls": 2, "tokens_in": 15435, "tokens_out": 3176, "ctx_chars": 11823, "t_end": 1791534849.986235, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0201, "lo95": 1.0181, "mean": 1.0189, "ev_only": 1.0202, "ref_ms": 27.58779, "cand_ms": 27.04381, "n_trials": 28, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.015, "r_wall_s": 57.05, "compile_s": 1.89, "r_error": "", "src_len": 1979, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 2136, "ts": 1791534681.435478, "chk": "ok", "code_len": 1979}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8512, "tout": 1040, "ts": 1791534792.8858988, "chk": null, "code_len": 1979}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791534661.8039796}
{"ep": "h2-00386", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.6413, "speedup": 1.6413, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 216.1, "calls": 2, "tokens_in": 7701, "tokens_out": 12170, "ctx_chars": 0, "t_end": 1791534850.0939336, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.6413, "lo95": 1.6401, "mean": 1.6417, "ev_only": 1.6436, "ref_ms": 7.77922, "cand_ms": 4.73964, "n_trials": 162, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 81.93, "compile_s": 10.75, "r_error": "", "src_len": 5182, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 9209, "ts": 1791534722.0056071, "chk": "ok", "code_len": 5182}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5542, "tout": 2961, "ts": 1791534767.4876354, "chk": null, "code_len": 5182}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2159, "started": 1791534634.0073981}
{"ep": "h2-00424", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.2994, "speedup": 2.2994, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 91.5, "calls": 2, "tokens_in": 13659, "tokens_out": 4309, "ctx_chars": 8529, "t_end": 1791534852.4940858, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.2994, "lo95": 2.2769, "mean": 2.2886, "ev_only": 2.3075, "ref_ms": 8.5465, "cand_ms": 3.71687, "n_trials": 79, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.513, "r_wall_s": 41.07, "compile_s": 1.87, "r_error": "", "src_len": 3398, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2557, "ts": 1791534785.9060087, "chk": "ok", "code_len": 3455}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8164, "tout": 1752, "ts": 1791534811.3325908, "chk": null, "code_len": 3398}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791534761.0010574}
{"ep": "h2-00368", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0489, "speedup": 1.0489, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 284.7, "calls": 3, "tokens_in": 18617, "tokens_out": 3743, "ctx_chars": 6513, "t_end": 1791534852.5457141, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0489, "lo95": 1.0429, "mean": 1.0461, "ev_only": 1.0497, "ref_ms": 15.81322, "cand_ms": 15.07611, "n_trials": 51, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1717, "r_wall_s": 195.35, "compile_s": 8.48, "r_error": "", "src_len": 3793, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1879, "ts": 1791534583.400541, "chk": "ok", "code_len": 3793}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6853, "tout": 25, "ts": 1791534640.7333465, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6902, "tout": 1839, "ts": 1791534656.826053, "chk": null, "code_len": 3793}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791534567.874332}
{"ep": "h2-00437", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0075, "speedup": 1.0075, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 40.1, "calls": 1, "tokens_in": 6367, "tokens_out": 531, "ctx_chars": 10847, "t_end": 1791534856.2461112, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0075, "lo95": 1.0063, "mean": 1.044, "ev_only": 1.0081, "ref_ms": 7.97186, "cand_ms": 7.91231, "n_trials": 128, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 34.79, "compile_s": 1.88, "r_error": "", "src_len": 993, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 531, "ts": 1791534821.3423347, "chk": null, "code_len": 993}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791534816.139887}
{"ep": "h2-00394", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9231, "speedup": 0.9231, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 213.2, "calls": 3, "tokens_in": 23988, "tokens_out": 2055, "ctx_chars": 11539, "t_end": 1791534858.588036, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9231, "lo95": 0.9226, "mean": 0.9246, "ev_only": 0.9238, "ref_ms": 7.75563, "cand_ms": 8.40197, "n_trials": 148, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0768, "r_wall_s": 58.57, "compile_s": 1.89, "r_error": "", "src_len": 1810, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1197, "ts": 1791534658.4100044, "chk": "ok", "code_len": 1810}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8419, "tout": 13, "ts": 1791534792.3071535, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8456, "tout": 845, "ts": 1791534799.3713906, "chk": null, "code_len": 1810}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791534645.4130065}
{"ep": "h2-00427", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7779, "speedup": 0.7779, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 86.7, "calls": 3, "tokens_in": 21108, "tokens_out": 950, "ctx_chars": 10850, "t_end": 1791534861.6310134, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.7779, "lo95": 0.7622, "mean": 0.7132, "ev_only": 0.7735, "ref_ms": 0.1485, "cand_ms": 0.1909, "n_trials": 206, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 67.6, "compile_s": 2.53, "r_error": "", "src_len": 804, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 449, "ts": 1791534780.0538611, "chk": "ok", "code_len": 804}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7180, "tout": 102, "ts": 1791534789.733511, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7306, "tout": 399, "ts": 1791534793.79807, "chk": null, "code_len": 804}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6622, "started": 1791534774.8968878}
{"ep": "h2-00425", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.002, "speedup": 1.002, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 103.3, "calls": 3, "tokens_in": 20258, "tokens_out": 2919, "ctx_chars": 9266, "t_end": 1791534864.7816317, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.002, "lo95": 0.9993, "mean": 1.0012, "ev_only": 1.0022, "ref_ms": 30.42687, "cand_ms": 30.36664, "n_trials": 33, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0383, "r_wall_s": 56.26, "compile_s": 1.9, "r_error": "", "src_len": 2724, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1571, "ts": 1791534777.6694107, "chk": "ok", "code_len": 2724}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7297, "tout": 26, "ts": 1791534796.932161, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7347, "tout": 1322, "ts": 1791534807.743936, "chk": null, "code_len": 2724}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791534761.45862}
{"ep": "h2-00374", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0279, "speedup": 1.0279, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 266.5, "calls": 2, "tokens_in": 12844, "tokens_out": 1733, "ctx_chars": 9656, "t_end": 1791534868.9655492, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0279, "lo95": 1.0265, "mean": 1.0279, "ev_only": 1.0276, "ref_ms": 24.96852, "cand_ms": 24.29005, "n_trials": 54, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0287, "r_wall_s": 207.54, "compile_s": 11.79, "r_error": "", "src_len": 1043, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 1064, "ts": 1791534612.049415, "chk": "ok", "code_len": 1043}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7010, "tout": 669, "ts": 1791534660.7369227, "chk": null, "code_len": 1043}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791534602.457358}
{"ep": "h2-00339", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8587, "speedup": 0.8587, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 389.0, "calls": 5, "tokens_in": 58295, "tokens_out": 12801, "ctx_chars": 6513, "t_end": 1791534869.1548867, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.8587, "lo95": 0.7961, "mean": 0.8232, "ev_only": 0.8584, "ref_ms": 8.00749, "cand_ms": 9.3249, "n_trials": 121, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0007, "r_wall_s": 52.76, "compile_s": 1.96, "r_error": "", "src_len": 3712, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2763, "ts": 1791534503.9500844, "chk": "incorrect", "code_len": 6358}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8540, "tout": 3474, "ts": 1791534602.1263466, "chk": "incorrect", "code_len": 6426}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 12146, "tout": 2447, "ts": 1791534636.3334875, "chk": "incorrect", "code_len": 4812}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 14725, "tout": 2382, "ts": 1791534693.2996619, "chk": "ok", "code_len": 3712}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 17239, "tout": 1735, "ts": 1791534816.3728364, "chk": null, "code_len": 3712}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "incorrect", "ok", null], "first_tin": 5645, "started": 1791534480.1259718}
{"ep": "h2-00435", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0007, "speedup": 1.0007, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 55.9, "calls": 1, "tokens_in": 6477, "tokens_out": 496, "ctx_chars": 10909, "t_end": 1791534869.2001722, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0007, "lo95": 0.9995, "mean": 0.9989, "ev_only": 1.0016, "ref_ms": 7.98625, "cand_ms": 7.9804, "n_trials": 187, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0024, "r_wall_s": 46.77, "compile_s": 2.99, "r_error": "", "src_len": 943, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 496, "ts": 1791534822.0262835, "chk": null, "code_len": 943}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791534813.2717857}
{"ep": "h2-00416", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.5208, "speedup": 1.5208, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 159.4, "calls": 2, "tokens_in": 16777, "tokens_out": 6790, "ctx_chars": 11539, "t_end": 1791534869.2245526, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.5208, "lo95": 1.5201, "mean": 1.5211, "ev_only": 1.5209, "ref_ms": 31.43966, "cand_ms": 20.67323, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9562, "r_wall_s": 85.84, "compile_s": 2.01, "r_error": "", "src_len": 4359, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 4487, "ts": 1791534748.1361423, "chk": "ok", "code_len": 4359}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9743, "tout": 2303, "ts": 1791534782.562927, "chk": null, "code_len": 4359}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791534710.3440347}
{"ep": "h2-00409", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 7.5602, "speedup": 7.5602, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 189.2, "calls": 2, "tokens_in": 6655, "tokens_out": 14152, "ctx_chars": 0, "t_end": 1791534870.333891, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 7.5602, "lo95": 7.4754, "mean": 7.3543, "ev_only": 7.5801, "ref_ms": 64.70279, "cand_ms": 8.55829, "n_trials": 20, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0206, "r_wall_s": 51.06, "compile_s": 2.53, "r_error": "", "src_len": 4215, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 11936, "ts": 1791534795.9881058, "chk": "ok", "code_len": 4442}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4909, "tout": 2216, "ts": 1791534819.2552998, "chk": null, "code_len": 4215}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791534681.1685312}
{"ep": "h2-00309", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0546, "speedup": 1.0546, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 497.0, "calls": 3, "tokens_in": 22794, "tokens_out": 2375, "ctx_chars": 10527, "t_end": 1791534875.5685782, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0546, "lo95": 1.054, "mean": 1.0551, "ev_only": 1.0555, "ref_ms": 19.62219, "cand_ms": 18.6067, "n_trials": 66, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0043, "r_wall_s": 225.19, "compile_s": 3.44, "r_error": "", "src_len": 2154, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1247, "ts": 1791534390.0590603, "chk": "ok", "code_len": 2283}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8020, "tout": 69, "ts": 1791534641.2874632, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8113, "tout": 1059, "ts": 1791534650.3416498, "chk": null, "code_len": 2154}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791534378.5762486}
{"ep": "h2-00410", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9247, "speedup": 0.9247, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 203.8, "calls": 3, "tokens_in": 24120, "tokens_out": 2191, "ctx_chars": 11539, "t_end": 1791534885.004364, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.9247, "lo95": 0.9242, "mean": 0.9246, "ev_only": 0.9234, "ref_ms": 7.76544, "cand_ms": 8.39808, "n_trials": 165, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0765, "r_wall_s": 62.5, "compile_s": 2.76, "r_error": "", "src_len": 1814, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1193, "ts": 1791534693.639438, "chk": "ok", "code_len": 1814}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8415, "tout": 153, "ts": 1791534812.5494585, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8592, "tout": 845, "ts": 1791534821.845518, "chk": null, "code_len": 1814}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791534681.198202}
{"ep": "h2-00403", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0066, "speedup": 1.0066, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 226.4, "calls": 1, "tokens_in": 1623, "tokens_out": 2277, "ctx_chars": 0, "t_end": 1791534886.782191, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0066, "lo95": 1.0049, "mean": 1.0067, "ev_only": 1.0061, "ref_ms": 30.64835, "cand_ms": 30.44869, "n_trials": 19, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0423, "r_wall_s": 205.16, "compile_s": 72.82, "r_error": "", "src_len": 4876, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1623, "tout": 2277, "ts": 1791534681.176287, "chk": null, "code_len": 4876}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1623, "started": 1791534660.4152107}
{"ep": "h2-00384", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0165, "speedup": 1.0165, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 259.2, "calls": 3, "tokens_in": 15046, "tokens_out": 2459, "ctx_chars": 5026, "t_end": 1791534889.2547333, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0165, "lo95": 1.0158, "mean": 1.0173, "ev_only": 1.0185, "ref_ms": 8.14745, "cand_ms": 8.01508, "n_trials": 129, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.009, "r_wall_s": 24.24, "compile_s": 2.4, "r_error": "", "src_len": 1995, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1390, "ts": 1791534651.842968, "chk": "ok", "code_len": 1995}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5495, "tout": 39, "ts": 1791534855.0729692, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5558, "tout": 1030, "ts": 1791534864.7157512, "chk": null, "code_len": 1995}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 3993, "started": 1791534630.0180273}
{"ep": "h2-00396", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.062, "speedup": 1.062, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 242.5, "calls": 1, "tokens_in": 1633, "tokens_out": 3791, "ctx_chars": 0, "t_end": 1791534892.476615, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.062, "lo95": 1.0613, "mean": 1.0688, "ev_only": 1.06, "ref_ms": 8.487, "cand_ms": 7.99134, "n_trials": 78, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0022, "r_wall_s": 205.35, "compile_s": 9.5, "r_error": "", "src_len": 1599, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1633, "tout": 3791, "ts": 1791534686.9026926, "chk": null, "code_len": 1599}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1633, "started": 1791534650.024049}
{"ep": "h2-00355", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 36.9905, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 360.7, "calls": 2, "tokens_in": 7204, "tokens_out": 7927, "ctx_chars": 0, "t_end": 1791534892.5844405, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 36.9905, "lo95": 36.6501, "mean": 36.087, "ev_only": 36.9832, "ref_ms": 19.70581, "cand_ms": 0.53273, "n_trials": 56, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 238.44, "compile_s": 21.08, "r_error": "", "src_len": 5411, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 5378, "ts": 1791534578.565322, "chk": "ok", "code_len": 5411}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5246, "tout": 2549, "ts": 1791534654.096888, "chk": null, "code_len": 5411}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791534531.934404}
{"ep": "h2-00367", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 5.1306, "speedup": 5.1306, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 337.8, "calls": 2, "tokens_in": 6569, "tokens_out": 7554, "ctx_chars": 0, "t_end": 1791534905.1195333, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 5.1306, "lo95": 4.9682, "mean": 5.3223, "ev_only": 5.1338, "ref_ms": 1.8857, "cand_ms": 0.36754, "n_trials": 164, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 259.28, "compile_s": 16.55, "r_error": "", "src_len": 5098, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4913, "ts": 1791534611.026439, "chk": "ok", "code_len": 5155}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4810, "tout": 2641, "ts": 1791534645.8137872, "chk": null, "code_len": 5098}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1759, "started": 1791534567.2935147}
{"ep": "h2-00440", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0587, "speedup": 1.0587, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 74.9, "calls": 2, "tokens_in": 14718, "tokens_out": 2348, "ctx_chars": 10527, "t_end": 1791534907.5333228, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0587, "lo95": 1.0579, "mean": 1.0555, "ev_only": 1.0583, "ref_ms": 19.79784, "cand_ms": 18.6999, "n_trials": 37, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0085, "r_wall_s": 43.37, "compile_s": 2.91, "r_error": "", "src_len": 2149, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1284, "ts": 1791534846.110939, "chk": "ok", "code_len": 2297}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8057, "tout": 1064, "ts": 1791534864.128807, "chk": null, "code_len": 2149}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6661, "started": 1791534832.6425114}
{"ep": "h2-00430", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0064, "speedup": 1.0064, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 121.7, "calls": 2, "tokens_in": 14984, "tokens_out": 6074, "ctx_chars": 10823, "t_end": 1791534909.9955127, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0064, "lo95": 1.006, "mean": 1.0065, "ev_only": 1.0066, "ref_ms": 30.02482, "cand_ms": 29.8349, "n_trials": 51, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0213, "r_wall_s": 48.34, "compile_s": 1.87, "r_error": "", "src_len": 3399, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 4351, "ts": 1791534826.8542924, "chk": "ok", "code_len": 3399}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8629, "tout": 1723, "ts": 1791534861.1671507, "chk": null, "code_len": 3399}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791534788.2624226}
{"ep": "h2-00428", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0194, "speedup": 1.0194, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 130.0, "calls": 2, "tokens_in": 5564, "tokens_out": 5553, "ctx_chars": 0, "t_end": 1791534914.5103164, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0194, "lo95": 1.0191, "mean": 1.0198, "ev_only": 1.0197, "ref_ms": 30.47513, "cand_ms": 29.89576, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0195, "r_wall_s": 55.56, "compile_s": 2.54, "r_error": "", "src_len": 3583, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 3762, "ts": 1791534818.5149534, "chk": "ok", "code_len": 3583}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3939, "tout": 1791, "ts": 1791534858.1720827, "chk": null, "code_len": 3583}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791534784.559829}
{"ep": "h2-00382", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9985, "speedup": 0.9985, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 300.6, "calls": 4, "tokens_in": 25742, "tokens_out": 1582, "ctx_chars": 9033, "t_end": 1791534923.6548934, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9985, "lo95": 0.9957, "mean": 0.9979, "ev_only": 0.999, "ref_ms": 15.56154, "cand_ms": 15.58531, "n_trials": 99, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 60.78, "compile_s": 1.94, "r_error": "", "src_len": 493, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 685, "ts": 1791534631.3756242, "chk": "ok", "code_len": 732}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6282, "tout": 532, "ts": 1791534677.1234899, "chk": "ok", "code_len": 493}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6923, "tout": 102, "ts": 1791534859.0939603, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7049, "tout": 263, "ts": 1791534862.2132592, "chk": null, "code_len": 493}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 5488, "started": 1791534623.0744102}
{"ep": "h2-00436", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9753, "speedup": 0.9753, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 118.0, "calls": 1, "tokens_in": 7032, "tokens_out": 1247, "ctx_chars": 11539, "t_end": 1791534932.94979, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9753, "lo95": 0.975, "mean": 0.9757, "ev_only": 0.9756, "ref_ms": 31.36352, "cand_ms": 32.15625, "n_trials": 44, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0237, "r_wall_s": 102.02, "compile_s": 2.03, "r_error": "", "src_len": 1804, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1247, "ts": 1791534829.6288087, "chk": null, "code_len": 1804}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791534814.959313}
{"ep": "h2-00442", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9963, "speedup": 0.9963, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.0, "calls": 1, "tokens_in": 7032, "tokens_out": 1141, "ctx_chars": 11539, "t_end": 1791534935.0013993, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9963, "lo95": 0.9956, "mean": 0.9964, "ev_only": 0.9965, "ref_ms": 31.39009, "cand_ms": 31.50709, "n_trials": 51, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 79.19, "compile_s": 1.89, "r_error": "", "src_len": 1887, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1141, "ts": 1791534854.8227925, "chk": null, "code_len": 1887}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791534842.9968035}
{"ep": "h2-00377", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0132, "speedup": 1.0132, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 324.7, "calls": 2, "tokens_in": 17450, "tokens_out": 6047, "ctx_chars": 10167, "t_end": 1791534938.2071862, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0132, "lo95": 1.0127, "mean": 1.0323, "ev_only": 1.0137, "ref_ms": 8.44732, "cand_ms": 8.33758, "n_trials": 177, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0488, "r_wall_s": 26.27, "compile_s": 1.94, "r_error": "", "src_len": 1680, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 5130, "ts": 1791534660.4276915, "chk": "ok", "code_len": 1680}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 11346, "tout": 917, "ts": 1791534911.7925408, "chk": null, "code_len": 1680}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6104, "started": 1791534613.5429044}
{"ep": "h2-00445", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0226, "speedup": 1.0226, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 89.1, "calls": 2, "tokens_in": 9220, "tokens_out": 2442, "ctx_chars": 5026, "t_end": 1791534939.082468, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0226, "lo95": 1.022, "mean": 1.0627, "ev_only": 1.0234, "ref_ms": 8.16359, "cand_ms": 7.98349, "n_trials": 178, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0085, "r_wall_s": 44.62, "compile_s": 1.94, "r_error": "", "src_len": 1266, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1675, "ts": 1791534866.9389925, "chk": "ok", "code_len": 1266}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5225, "tout": 767, "ts": 1791534894.1612115, "chk": null, "code_len": 1266}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791534849.986485}
{"ep": "h2-00459", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0044, "speedup": 1.0044, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 57.0, "calls": 2, "tokens_in": 11404, "tokens_out": 1110, "ctx_chars": 9028, "t_end": 1791534941.997227, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0044, "lo95": 1.0017, "mean": 1.0154, "ev_only": 1.0041, "ref_ms": 8.04073, "cand_ms": 8.00549, "n_trials": 149, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 18.87, "compile_s": 1.88, "r_error": "", "src_len": 1041, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 554, "ts": 1791534890.2661843, "chk": "ok", "code_len": 1041}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6035, "tout": 556, "ts": 1791534923.0403798, "chk": null, "code_len": 1041}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791534885.0046275}
{"ep": "h2-00431", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9219, "speedup": 0.9219, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 146.7, "calls": 1, "tokens_in": 7113, "tokens_out": 1141, "ctx_chars": 11539, "t_end": 1791534942.288455, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9219, "lo95": 0.9212, "mean": 0.9218, "ev_only": 0.9214, "ref_ms": 7.76681, "cand_ms": 8.42473, "n_trials": 159, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0766, "r_wall_s": 134.35, "compile_s": 1.92, "r_error": "", "src_len": 1810, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7113, "tout": 1141, "ts": 1791534807.1893122, "chk": null, "code_len": 1810}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7113, "started": 1791534795.603207}
{"ep": "h2-00444", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0279, "speedup": 1.0279, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 93.3, "calls": 1, "tokens_in": 5832, "tokens_out": 526, "ctx_chars": 9656, "t_end": 1791534942.7378263, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0279, "lo95": 1.0252, "mean": 1.0267, "ev_only": 1.0278, "ref_ms": 24.97841, "cand_ms": 24.29932, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0223, "r_wall_s": 87.2, "compile_s": 1.84, "r_error": "", "src_len": 1002, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 526, "ts": 1791534854.422855, "chk": null, "code_len": 1002}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791534849.4805167}
{"ep": "h2-00454", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "runtime_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 76.2, "calls": 1, "tokens_in": 1964, "tokens_out": 6252, "ctx_chars": 0, "t_end": 1791534945.3209767, "error": "CompilationError: at 12:8:\ndef lstm_cell_kernel(xg_ptr, hg_ptr, c_ptr, cnew_ptr, hnew_ptr, H, BLOCK: tl.constexpr):\n    row = tl.program_id(0)\n    col = tl.prog", "f_status": "runtime_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "triton", "own_frac": null, "r_wall_s": 11.78, "compile_s": 1.95, "r_error": "CompilationError: at 12:8:\ndef lstm_cell_kernel(xg_ptr, hg_ptr, c_ptr, cnew_ptr, hnew_ptr, H, BLOCK: tl.constexpr):\n    row = tl.program_id(0)\n    col = tl.prog", "src_len": 2474, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1964, "tout": 6252, "ts": 1791534933.489931, "chk": null, "code_len": 2474}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1964, "started": 1791534869.1551673}
{"ep": "h2-00429", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9185, "speedup": 0.9185, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 160.7, "calls": 3, "tokens_in": 24357, "tokens_out": 2272, "ctx_chars": 11539, "t_end": 1791534946.111399, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9185, "lo95": 0.9176, "mean": 0.9195, "ev_only": 0.9213, "ref_ms": 7.80883, "cand_ms": 8.50185, "n_trials": 135, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0767, "r_wall_s": 98.9, "compile_s": 2.04, "r_error": "", "src_len": 1810, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1349, "ts": 1791534800.4113615, "chk": "ok", "code_len": 1810}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8571, "tout": 78, "ts": 1791534839.1064541, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8673, "tout": 845, "ts": 1791534846.518259, "chk": null, "code_len": 1810}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791534785.4305198}
{"ep": "h2-00451", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.005, "speedup": 1.005, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 89.6, "calls": 2, "tokens_in": 11396, "tokens_out": 1094, "ctx_chars": 9028, "t_end": 1791534951.202369, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.005, "lo95": 1.0047, "mean": 1.0468, "ev_only": 1.0038, "ref_ms": 8.00104, "cand_ms": 7.96115, "n_trials": 150, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 27.91, "compile_s": 2.01, "r_error": "", "src_len": 1024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 546, "ts": 1791534890.9956217, "chk": "ok", "code_len": 1024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6027, "tout": 548, "ts": 1791534923.1992717, "chk": null, "code_len": 1024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791534861.6312726}
{"ep": "h2-00397", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 10.0, "speedup": 14.9367, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 301.6, "calls": 2, "tokens_in": 15655, "tokens_out": 6770, "ctx_chars": 10527, "t_end": 1791534952.2599058, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 14.9367, "lo95": 14.8869, "mean": 14.849, "ev_only": 15.1075, "ref_ms": 19.6586, "cand_ms": 1.31613, "n_trials": 80, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "triton", "own_frac": 1.0, "r_wall_s": 25.3, "compile_s": 2.06, "r_error": "", "src_len": 3962, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 5006, "ts": 1791534697.523567, "chk": "ok", "code_len": 3991}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8992, "tout": 1764, "ts": 1791534926.8410933, "chk": null, "code_len": 3962}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791534650.7071788}
{"ep": "h2-00460", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0225, "speedup": 1.0225, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 67.1, "calls": 2, "tokens_in": 14647, "tokens_out": 2577, "ctx_chars": 11065, "t_end": 1791534953.8596303, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0225, "lo95": 1.0209, "mean": 1.0035, "ev_only": 1.0185, "ref_ms": 1.80332, "cand_ms": 1.76371, "n_trials": 216, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1128, "r_wall_s": 21.26, "compile_s": 1.86, "r_error": "", "src_len": 1570, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 1281, "ts": 1791534900.2131433, "chk": "ok", "code_len": 1602}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8020, "tout": 1296, "ts": 1791534932.4935327, "chk": null, "code_len": 1570}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791534886.7824934}
{"ep": "h2-00443", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.3618, "speedup": 2.3618, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 112.0, "calls": 3, "tokens_in": 20860, "tokens_out": 3961, "ctx_chars": 8529, "t_end": 1791534957.009485, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 2.3618, "lo95": 2.3574, "mean": 2.3647, "ev_only": 2.3655, "ref_ms": 8.58163, "cand_ms": 3.63355, "n_trials": 169, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4129, "r_wall_s": 65.69, "compile_s": 2.02, "r_error": "", "src_len": 3627, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2017, "ts": 1791534863.3564909, "chk": "ok", "code_len": 3663}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7624, "tout": 93, "ts": 1791534876.2435563, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7741, "tout": 1851, "ts": 1791534891.2589157, "chk": null, "code_len": 3627}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5495, "started": 1791534845.0319262}
{"ep": "h2-00349", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.2497, "speedup": 1.2497, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 456.8, "calls": 2, "tokens_in": 28941, "tokens_out": 18912, "ctx_chars": 0, "t_end": 1791534959.3309932, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.2497, "lo95": 1.1645, "mean": 1.6849, "ev_only": 1.2542, "ref_ms": 19.63388, "cand_ms": 15.71049, "n_trials": 67, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1216, "r_wall_s": 301.92, "compile_s": 8.69, "r_error": "", "src_len": 16303, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 11638, "ts": 1791534591.8251622, "chk": "ok", "code_len": 16303}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18307, "tout": 7274, "ts": 1791534657.375681, "chk": null, "code_len": 16303}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791534502.4957047}
{"ep": "h2-00356", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.0579, "speedup": 0.0579, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 434.2, "calls": 5, "tokens_in": 63298, "tokens_out": 12411, "ctx_chars": 8224, "t_end": 1791534968.3018441, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.0579, "lo95": 0.0531, "mean": 0.0567, "ev_only": 0.058, "ref_ms": 31.96845, "cand_ms": 551.67544, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1402, "r_wall_s": 66.17, "compile_s": 2.65, "r_error": "", "src_len": 3452, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 5608, "ts": 1791534600.2118518, "chk": "forbidden", "code_len": 3534}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 11277, "tout": 1875, "ts": 1791534617.0564914, "chk": "incorrect", "code_len": 3452}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 13284, "tout": 3168, "ts": 1791534670.093462, "chk": "ok", "code_len": 3496}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 16564, "tout": 54, "ts": 1791534888.1273627, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 16665, "tout": 1706, "ts": 1791534902.111841, "chk": null, "code_len": 3452}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "incorrect", "ok", "no_code", null], "first_tin": 5508, "started": 1791534534.072806}
{"ep": "h2-00407", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0167, "speedup": 1.0167, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 295.0, "calls": 1, "tokens_in": 6367, "tokens_out": 543, "ctx_chars": 10847, "t_end": 1791534969.4950528, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0167, "lo95": 1.012, "mean": 1.0158, "ev_only": 1.0117, "ref_ms": 8.11807, "cand_ms": 7.98457, "n_trials": 172, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 289.5, "compile_s": 13.04, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 543, "ts": 1791534679.8439016, "chk": null, "code_len": 999}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791534674.4657428}
{"ep": "h2-00449", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7988, "speedup": 0.7988, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 114.5, "calls": 2, "tokens_in": 12452, "tokens_out": 2103, "ctx_chars": 8224, "t_end": 1791534970.7580528, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7988, "lo95": 0.6746, "mean": 0.6917, "ev_only": 0.7982, "ref_ms": 60.27453, "cand_ms": 75.45558, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 77.43, "compile_s": 2.51, "r_error": "", "src_len": 1114, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 1287, "ts": 1791534872.9191, "chk": "ok", "code_len": 1114}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6924, "tout": 816, "ts": 1791534893.290738, "chk": null, "code_len": 1114}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5528, "started": 1791534856.2464073}
{"ep": "h2-00404", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 39.1082, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 309.1, "calls": 2, "tokens_in": 7031, "tokens_out": 7083, "ctx_chars": 0, "t_end": 1791534970.837035, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 39.1082, "lo95": 39.0311, "mean": 39.7795, "ev_only": 40.1053, "ref_ms": 19.66537, "cand_ms": 0.50285, "n_trials": 74, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 35.32, "compile_s": 13.06, "r_error": "", "src_len": 5193, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4624, "ts": 1791534700.5860763, "chk": "ok", "code_len": 5256}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5073, "tout": 2459, "ts": 1791534935.3983078, "chk": null, "code_len": 5193}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791534661.7641037}
{"ep": "h2-00392", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 8.1076, "speedup": 8.1076, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 332.9, "calls": 2, "tokens_in": 7010, "tokens_out": 19662, "ctx_chars": 0, "t_end": 1791534974.948235, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 8.1076, "lo95": 8.0477, "mean": 8.6365, "ev_only": 8.0972, "ref_ms": 64.37499, "cand_ms": 7.94006, "n_trials": 24, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0123, "r_wall_s": 75.11, "compile_s": 2.53, "r_error": "", "src_len": 5125, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 17021, "ts": 1791534802.9549549, "chk": "ok", "code_len": 5437}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5264, "tout": 2641, "ts": 1791534899.7919214, "chk": null, "code_len": 5125}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791534642.0341914}
{"ep": "h2-00398", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.3233, "speedup": 0.3233, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 327.6, "calls": 3, "tokens_in": 23171, "tokens_out": 6479, "ctx_chars": 6513, "t_end": 1791534978.5420127, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.3233, "lo95": 0.3173, "mean": 0.2945, "ev_only": 0.3237, "ref_ms": 3.32253, "cand_ms": 10.27559, "n_trials": 89, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0248, "r_wall_s": 55.77, "compile_s": 2.95, "r_error": "", "src_len": 6632, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 3271, "ts": 1791534676.9385443, "chk": "ok", "code_len": 6632}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8829, "tout": 43, "ts": 1791534884.9028323, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8896, "tout": 3165, "ts": 1791534922.7189033, "chk": null, "code_len": 6632}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791534650.8961158}
{"ep": "h2-00450", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9015, "speedup": 0.9015, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 120.2, "calls": 2, "tokens_in": 13832, "tokens_out": 899, "ctx_chars": 10850, "t_end": 1791534978.805267, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9015, "lo95": 0.8758, "mean": 0.8689, "ev_only": 0.8068, "ref_ms": 0.17943, "cand_ms": 0.19904, "n_trials": 197, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 94.17, "compile_s": 2.06, "r_error": "", "src_len": 807, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 479, "ts": 1791534864.066132, "chk": "ok", "code_len": 807}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7210, "tout": 420, "ts": 1791534884.490675, "chk": null, "code_len": 807}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6622, "started": 1791534858.5883005}
{"ep": "h2-00331", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.993, "speedup": 0.993, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 539.3, "calls": 3, "tokens_in": 23001, "tokens_out": 3321, "ctx_chars": 10823, "t_end": 1791534980.9743326, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.993, "lo95": 0.9925, "mean": 0.9926, "ev_only": 0.9934, "ref_ms": 30.23949, "cand_ms": 30.45152, "n_trials": 42, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0379, "r_wall_s": 325.1, "compile_s": 9.2, "r_error": "", "src_len": 3000, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1817, "ts": 1791534458.3764215, "chk": "ok", "code_len": 3546}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8282, "tout": 60, "ts": 1791534642.6158097, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8366, "tout": 1444, "ts": 1791534655.3429482, "chk": null, "code_len": 3000}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791534441.6252506}
{"ep": "h2-00389", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.3259, "speedup": 2.3259, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 341.6, "calls": 2, "tokens_in": 13162, "tokens_out": 3822, "ctx_chars": 8529, "t_end": 1791534982.2025454, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.3259, "lo95": 2.3208, "mean": 2.3131, "ev_only": 2.3302, "ref_ms": 8.49955, "cand_ms": 3.65428, "n_trials": 185, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4224, "r_wall_s": 53.93, "compile_s": 2.94, "r_error": "", "src_len": 3405, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2060, "ts": 1791534660.5473304, "chk": "ok", "code_len": 3405}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7667, "tout": 1762, "ts": 1791534928.2190275, "chk": null, "code_len": 3405}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791534640.5692556}
{"ep": "h2-00426", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4549, "speedup": 1.4549, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 208.3, "calls": 2, "tokens_in": 11168, "tokens_out": 4960, "ctx_chars": 6513, "t_end": 1791534982.8677845, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.4549, "lo95": 1.453, "mean": 1.4385, "ev_only": 1.4536, "ref_ms": 15.59368, "cand_ms": 10.71807, "n_trials": 87, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1835, "r_wall_s": 94.03, "compile_s": 2.02, "r_error": "", "src_len": 3236, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3455, "ts": 1791534806.5129101, "chk": "ok", "code_len": 3236}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6561, "tout": 1505, "ts": 1791534888.1465752, "chk": null, "code_len": 3236}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791534775.2869642}
{"ep": "h2-00401", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 36.9305, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 326.4, "calls": 2, "tokens_in": 7055, "tokens_out": 7432, "ctx_chars": 0, "t_end": 1791534984.1964917, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 36.9305, "lo95": 36.6314, "mean": 35.6181, "ev_only": 37.9664, "ref_ms": 19.71964, "cand_ms": 0.53397, "n_trials": 67, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 57.23, "compile_s": 12.2, "r_error": "", "src_len": 5351, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4852, "ts": 1791534698.850529, "chk": "ok", "code_len": 5351}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5097, "tout": 2580, "ts": 1791534926.906366, "chk": null, "code_len": 5351}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791534657.7719843}
{"ep": "h2-00373", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.7823, "speedup": 2.7823, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 386.5, "calls": 2, "tokens_in": 13758, "tokens_out": 4379, "ctx_chars": 9970, "t_end": 1791534986.6062043, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.7823, "lo95": 2.7793, "mean": 2.7759, "ev_only": 2.7935, "ref_ms": 7.82592, "cand_ms": 2.81277, "n_trials": 220, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 328.19, "compile_s": 11.03, "r_error": "", "src_len": 2571, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 3109, "ts": 1791534628.2424889, "chk": "ok", "code_len": 2571}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7758, "tout": 1270, "ts": 1791534657.6463742, "chk": null, "code_len": 2571}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791534600.1347904}
{"ep": "h2-00452", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0614, "speedup": 1.0614, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.9, "calls": 2, "tokens_in": 4959, "tokens_out": 3474, "ctx_chars": 0, "t_end": 1791534990.6717858, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0614, "lo95": 1.0596, "mean": 1.062, "ev_only": 1.0601, "ref_ms": 8.47609, "cand_ms": 7.98564, "n_trials": 174, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0053, "r_wall_s": 58.99, "compile_s": 3.27, "r_error": "", "src_len": 2339, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 2185, "ts": 1791534882.6581993, "chk": "ok", "code_len": 2339}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3324, "tout": 1289, "ts": 1791534931.4821546, "chk": null, "code_len": 2339}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791534864.7818909}
{"ep": "h2-00432", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.405, "speedup": 1.405, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 196.7, "calls": 2, "tokens_in": 16784, "tokens_out": 8261, "ctx_chars": 11539, "t_end": 1791534995.6079493, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.405, "lo95": 1.4045, "mean": 1.4045, "ev_only": 1.4055, "ref_ms": 31.39852, "cand_ms": 22.34693, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9594, "r_wall_s": 96.36, "compile_s": 2.14, "r_error": "", "src_len": 4264, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 6056, "ts": 1791534851.4946465, "chk": "ok", "code_len": 4264}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9750, "tout": 2205, "ts": 1791534898.415058, "chk": null, "code_len": 4264}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791534798.8910658}
{"ep": "h2-00448", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.07, "speedup": 0.07, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 143.1, "calls": 3, "tokens_in": 29345, "tokens_out": 8053, "ctx_chars": 8224, "t_end": 1791534995.6102037, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.07, "lo95": 0.0688, "mean": 0.0717, "ev_only": 0.07, "ref_ms": 61.27594, "cand_ms": 875.60606, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.2181, "r_wall_s": 59.66, "compile_s": 2.06, "r_error": "", "src_len": 3414, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 6149, "ts": 1791534905.4801726, "chk": "ok", "code_len": 3414}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 11789, "tout": 215, "ts": 1791534921.4863899, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 12028, "tout": 1689, "ts": 1791534935.8910837, "chk": null, "code_len": 3414}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791534852.5459847}
{"ep": "h2-00446", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.0917, "speedup": 3.0917, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 147.7, "calls": 2, "tokens_in": 5856, "tokens_out": 7919, "ctx_chars": 0, "t_end": 1791534997.7870471, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 3.0917, "lo95": 3.0879, "mean": 3.118, "ev_only": 3.1, "ref_ms": 27.55549, "cand_ms": 8.91267, "n_trials": 36, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9212, "r_wall_s": 61.88, "compile_s": 2.0, "r_error": "", "src_len": 3221, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 6142, "ts": 1791534907.7345817, "chk": "ok", "code_len": 3221}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4126, "tout": 1777, "ts": 1791534935.8535407, "chk": null, "code_len": 3221}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791534850.0941527}
{"ep": "h2-00477", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7401, "speedup": 0.7401, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 52.4, "calls": 1, "tokens_in": 6622, "tokens_out": 508, "ctx_chars": 10850, "t_end": 1791534998.4881063, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.7401, "lo95": 0.7364, "mean": 0.7372, "ev_only": 0.7401, "ref_ms": 0.14534, "cand_ms": 0.19638, "n_trials": 185, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 46.69, "compile_s": 2.11, "r_error": "", "src_len": 870, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6622, "tout": 508, "ts": 1791534951.599525, "chk": null, "code_len": 870}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6622, "started": 1791534946.1116455}
{"ep": "h2-00461", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0198, "speedup": 1.0198, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.4, "calls": 3, "tokens_in": 15120, "tokens_out": 2555, "ctx_chars": 5026, "t_end": 1791535000.6613467, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0198, "lo95": 1.0174, "mean": 1.0187, "ev_only": 1.0202, "ref_ms": 8.26647, "cand_ms": 8.10584, "n_trials": 192, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0114, "r_wall_s": 47.68, "compile_s": 2.46, "r_error": "", "src_len": 2095, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1418, "ts": 1791534902.7156894, "chk": "ok", "code_len": 2095}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5523, "tout": 57, "ts": 1791534941.2354674, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5604, "tout": 1080, "ts": 1791534952.883508, "chk": null, "code_len": 2095}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 3993, "started": 1791534889.255025}
{"ep": "h2-00415", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 3.3526, "speedup": 3.3526, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 295.6, "calls": 3, "tokens_in": 17101, "tokens_out": 16574, "ctx_chars": 0, "t_end": 1791535001.2739866, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 3.3526, "lo95": 3.038, "mean": 3.5122, "ev_only": 3.36, "ref_ms": 88.37474, "cand_ms": 26.35996, "n_trials": 13, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9065, "r_wall_s": 31.35, "compile_s": 11.46, "r_error": "", "src_len": 6759, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 9440, "ts": 1791534787.8894901, "chk": "no_custom_kernel", "code_len": 6746}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5831, "tout": 3830, "ts": 1791534851.9564645, "chk": "ok", "code_len": 6759}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9284, "tout": 3304, "ts": 1791534969.4529588, "chk": null, "code_len": 6759}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791534705.6721644}
{"ep": "h2-00478", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.997, "speedup": 0.997, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 50.1, "calls": 2, "tokens_in": 13450, "tokens_out": 1081, "ctx_chars": 10847, "t_end": 1791535001.3395143, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.997, "lo95": 0.9963, "mean": 0.9932, "ev_only": 1.0002, "ref_ms": 7.9596, "cand_ms": 7.98345, "n_trials": 190, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0043, "r_wall_s": 27.53, "compile_s": 1.98, "r_error": "", "src_len": 821, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 603, "ts": 1791534957.1688583, "chk": "ok", "code_len": 821}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7081, "tout": 478, "ts": 1791534973.4732578, "chk": null, "code_len": 821}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791534951.2026224}
{"ep": "h2-00438", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.6743, "speedup": 0.6743, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 183.3, "calls": 3, "tokens_in": 50025, "tokens_out": 14949, "ctx_chars": 2425, "t_end": 1791535002.866107, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.6743, "lo95": 0.5668, "mean": 0.742, "ev_only": 0.6751, "ref_ms": 20.74788, "cand_ms": 30.76839, "n_trials": 36, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0299, "r_wall_s": 25.91, "compile_s": 2.06, "r_error": "", "src_len": 17576, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7545, "ts": 1791534879.3116267, "chk": "ok", "code_len": 17576}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19205, "tout": 43, "ts": 1791534918.4288323, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19272, "tout": 7361, "ts": 1791534976.5819473, "chk": null, "code_len": 17576}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791534819.5961063}
{"ep": "h2-00250", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 5.3079, "speedup": 5.3079, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 795.6, "calls": 3, "tokens_in": 13585, "tokens_out": 9649, "ctx_chars": 0, "t_end": 1791535005.831944, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 5.3079, "lo95": 5.2708, "mean": 5.0468, "ev_only": 5.2853, "ref_ms": 1.81013, "cand_ms": 0.34102, "n_trials": 191, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 343.32, "compile_s": 27.94, "r_error": "", "src_len": 4776, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4674, "ts": 1791534250.4630432, "chk": "no_custom_kernel", "code_len": 4798}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4643, "tout": 2539, "ts": 1791534388.0832283, "chk": "ok", "code_len": 4776}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7183, "tout": 2436, "ts": 1791534662.484248, "chk": null, "code_len": 4776}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1759, "started": 1791534210.2020164}
{"ep": "h2-00447", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.0075, "speedup": 3.0075, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.7, "calls": 2, "tokens_in": 6544, "tokens_out": 7191, "ctx_chars": 0, "t_end": 1791535007.215925, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 3.0075, "lo95": 3.0056, "mean": 3.0037, "ev_only": 3.0063, "ref_ms": 42.54309, "cand_ms": 14.14584, "n_trials": 39, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.8973, "r_wall_s": 70.7, "compile_s": 3.27, "r_error": "", "src_len": 3566, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 5354, "ts": 1791534905.7126122, "chk": "ok", "code_len": 3566}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4508, "tout": 1837, "ts": 1791534935.7688196, "chk": null, "code_len": 3566}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2036, "started": 1791534852.4943476}
{"ep": "h2-00406", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5495, "speedup": 1.5495, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 341.8, "calls": 3, "tokens_in": 16242, "tokens_out": 12928, "ctx_chars": 0, "t_end": 1791535007.861391, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.5495, "lo95": 1.5486, "mean": 1.5515, "ev_only": 1.5473, "ref_ms": 7.7778, "cand_ms": 5.01939, "n_trials": 183, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 133.11, "compile_s": 13.93, "r_error": "", "src_len": 5588, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6634, "ts": 1791534724.0335555, "chk": "no_custom_kernel", "code_len": 5591}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5547, "tout": 3512, "ts": 1791534777.3135095, "chk": "ok", "code_len": 5588}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8536, "tout": 2782, "ts": 1791534874.0587332, "chk": null, "code_len": 5588}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791534666.0510542}
{"ep": "h2-00387", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 5.295, "speedup": 5.295, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 374.0, "calls": 3, "tokens_in": 33562, "tokens_out": 28082, "ctx_chars": 8224, "t_end": 1791535010.7673645, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 5.295, "lo95": 5.2816, "mean": 5.2357, "ev_only": 5.3074, "ref_ms": 32.19852, "cand_ms": 6.0809, "n_trials": 33, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9998, "r_wall_s": 54.04, "compile_s": 11.22, "r_error": "", "src_len": 10600, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 16930, "ts": 1791534792.2038422, "chk": "no_custom_kernel", "code_len": 10525}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 11315, "tout": 6041, "ts": 1791534867.0779846, "chk": "ok", "code_len": 10600}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 16737, "tout": 5111, "ts": 1791534956.6861007, "chk": null, "code_len": 10600}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791534636.8044941}
{"ep": "h2-00379", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 25.7059, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 396.3, "calls": 3, "tokens_in": 15155, "tokens_out": 10138, "ctx_chars": 0, "t_end": 1791535016.1352882, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 25.7059, "lo95": 25.49, "mean": 26.3593, "ev_only": 26.552, "ref_ms": 19.66069, "cand_ms": 0.76483, "n_trials": 61, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 60.22, "compile_s": 16.65, "r_error": "", "src_len": 5060, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 5021, "ts": 1791534662.4621646, "chk": "no_custom_kernel", "code_len": 5082}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5273, "tout": 2607, "ts": 1791534914.4565232, "chk": "ok", "code_len": 5060}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7924, "tout": 2510, "ts": 1791534955.8791256, "chk": null, "code_len": 5060}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1958, "started": 1791534619.799492}
{"ep": "h2-00458", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.2138, "speedup": 1.2138, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 141.1, "calls": 2, "tokens_in": 6498, "tokens_out": 6056, "ctx_chars": 0, "t_end": 1791535016.6995087, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.2138, "lo95": 1.2127, "mean": 1.2137, "ev_only": 1.2132, "ref_ms": 31.41066, "cand_ms": 25.87782, "n_trials": 42, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9998, "r_wall_s": 72.85, "compile_s": 2.2, "r_error": "", "src_len": 3147, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 4326, "ts": 1791534911.6131976, "chk": "ok", "code_len": 3147}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4420, "tout": 1730, "ts": 1791534942.7456548, "chk": null, "code_len": 3147}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791534875.5688891}
{"ep": "h2-00357", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.1388, "speedup": 1.1388, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 476.2, "calls": 2, "tokens_in": 31725, "tokens_out": 18879, "ctx_chars": 2425, "t_end": 1791535017.5537236, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.1388, "lo95": 1.0878, "mean": 1.1243, "ev_only": 1.14, "ref_ms": 19.67785, "cand_ms": 17.27883, "n_trials": 39, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1082, "r_wall_s": 66.49, "compile_s": 49.02, "r_error": "", "src_len": 18958, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 10735, "ts": 1791534619.5703015, "chk": "ok", "code_len": 18958}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 20175, "tout": 8144, "ts": 1791534951.0243647, "chk": null, "code_len": 18958}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791534541.3132741}
{"ep": "h2-00366", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 4.7152, "speedup": 4.7152, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 455.9, "calls": 3, "tokens_in": 27089, "tokens_out": 14462, "ctx_chars": 8224, "t_end": 1791535018.5852501, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 4.7152, "lo95": 4.0378, "mean": 4.4185, "ev_only": 4.723, "ref_ms": 85.10092, "cand_ms": 18.04809, "n_trials": 12, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8626, "r_wall_s": 93.7, "compile_s": 16.22, "r_error": "", "src_len": 6138, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 7985, "ts": 1791534631.970524, "chk": "no_custom_kernel", "code_len": 6160}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9177, "tout": 3259, "ts": 1791534670.2631762, "chk": "ok", "code_len": 6138}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 12382, "tout": 3218, "ts": 1791534924.8060093, "chk": null, "code_len": 6138}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791534562.684859}
{"ep": "h2-00455", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0667, "speedup": 1.0667, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 150.8, "calls": 3, "tokens_in": 16407, "tokens_out": 2307, "ctx_chars": 6513, "t_end": 1791535020.032495, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0667, "lo95": 1.0664, "mean": 1.083, "ev_only": 1.0657, "ref_ms": 15.47911, "cand_ms": 14.5108, "n_trials": 102, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1326, "r_wall_s": 77.09, "compile_s": 2.06, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1148, "ts": 1791534879.2109435, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5865, "tout": 48, "ts": 1791534932.7253146, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5937, "tout": 1111, "ts": 1791534942.2653959, "chk": null, "code_len": 2422}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4605, "started": 1791534869.2003636}
{"ep": "h2-00453", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0527, "speedup": 1.0527, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 153.7, "calls": 2, "tokens_in": 11975, "tokens_out": 4228, "ctx_chars": 6513, "t_end": 1791535022.6615376, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0527, "lo95": 1.0524, "mean": 1.0525, "ev_only": 1.0526, "ref_ms": 15.69605, "cand_ms": 14.91062, "n_trials": 99, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1722, "r_wall_s": 77.48, "compile_s": 2.05, "r_error": "", "src_len": 4295, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 2139, "ts": 1791534902.8293946, "chk": "ok", "code_len": 4295}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7113, "tout": 2089, "ts": 1791534944.9626436, "chk": null, "code_len": 4295}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4862, "started": 1791534868.9658463}
{"ep": "h2-00388", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0096, "speedup": 1.0096, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 388.8, "calls": 3, "tokens_in": 24641, "tokens_out": 2141, "ctx_chars": 12117, "t_end": 1791535026.4987001, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0096, "lo95": 1.0085, "mean": 1.009, "ev_only": 1.0096, "ref_ms": 42.66475, "cand_ms": 42.25821, "n_trials": 35, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 339.9, "compile_s": 9.25, "r_error": "", "src_len": 2251, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1075, "ts": 1791534648.9912035, "chk": "ok", "code_len": 2251}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8584, "tout": 52, "ts": 1791534677.1818788, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8660, "tout": 1014, "ts": 1791534685.9132938, "chk": null, "code_len": 2251}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791534637.6557996}
{"ep": "h2-00465", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.042, "speedup": 1.042, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 129.9, "calls": 2, "tokens_in": 4874, "tokens_out": 2690, "ctx_chars": 0, "t_end": 1791535037.403879, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.042, "lo95": 1.0414, "mean": 1.0574, "ev_only": 1.0429, "ref_ms": 8.28404, "cand_ms": 7.95019, "n_trials": 217, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0052, "r_wall_s": 94.46, "compile_s": 2.43, "r_error": "", "src_len": 1680, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 1765, "ts": 1791534923.4497812, "chk": "ok", "code_len": 1680}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3108, "tout": 925, "ts": 1791534942.8509035, "chk": null, "code_len": 1680}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791534907.5335999}
{"ep": "h2-00408", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 4.2083, "speedup": 4.2083, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 357.3, "calls": 3, "tokens_in": 31488, "tokens_out": 27331, "ctx_chars": 8224, "t_end": 1791535037.5545387, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 4.2083, "lo95": 4.1954, "mean": 4.3978, "ev_only": 4.2065, "ref_ms": 32.78189, "cand_ms": 7.78987, "n_trials": 46, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9865, "r_wall_s": 83.24, "compile_s": 12.69, "r_error": "", "src_len": 9394, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 17602, "ts": 1791534837.4191327, "chk": "no_custom_kernel", "code_len": 9378}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 10604, "tout": 5280, "ts": 1791534896.6412613, "chk": "ok", "code_len": 9394}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 15374, "tout": 4449, "ts": 1791534954.2660534, "chk": null, "code_len": 9394}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791534680.2140872}
{"ep": "h2-00489", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 78.8, "calls": 1, "tokens_in": 1723, "tokens_out": 1076, "ctx_chars": 0, "t_end": 1791535057.5893996, "error": "RuntimeError: Error building extension 'fused_conv3d_act': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_conv3d_act -DTORCH_API_INCLUDE_EXTENSION_H -", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 69.13, "compile_s": 69.12, "r_error": "RuntimeError: Error building extension 'fused_conv3d_act': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_conv3d_act -DTORCH_API_INCLUDE_EXTENSION_H -", "src_len": 2457, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1723, "tout": 1076, "ts": 1791534988.213563, "chk": null, "code_len": 2457}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1723, "started": 1791534978.8055146}
{"ep": "h2-00434", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7704, "speedup": 0.7704, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 257.2, "calls": 3, "tokens_in": 49830, "tokens_out": 14790, "ctx_chars": 2425, "t_end": 1791535067.2367172, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.7704, "lo95": 0.7682, "mean": 0.778, "ev_only": 0.7711, "ref_ms": 18.69517, "cand_ms": 24.26709, "n_trials": 60, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0299, "r_wall_s": 24.42, "compile_s": 1.85, "r_error": "", "src_len": 17399, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7453, "ts": 1791534886.733811, "chk": "ok", "code_len": 17399}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19113, "tout": 32, "ts": 1791534968.5364711, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19169, "tout": 7305, "ts": 1791535042.7580268, "chk": null, "code_len": 17399}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791534809.9883046}
{"ep": "h2-00491", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.3292, "speedup": 0.3292, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 85.8, "calls": 3, "tokens_in": 22626, "tokens_out": 5965, "ctx_chars": 6513, "t_end": 1791535068.032277, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.3292, "lo95": 0.3281, "mean": 0.3304, "ev_only": 0.3277, "ref_ms": 3.21332, "cand_ms": 9.76045, "n_trials": 147, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0249, "r_wall_s": 13.22, "compile_s": 1.92, "r_error": "", "src_len": 6126, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 3007, "ts": 1791535007.2164872, "chk": "ok", "code_len": 6126}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8565, "tout": 26, "ts": 1791535031.9040222, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8615, "tout": 2932, "ts": 1791535054.697218, "chk": null, "code_len": 6126}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791534982.20286}
{"ep": "h2-00457", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.965, "speedup": 0.965, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 199.1, "calls": 2, "tokens_in": 4847, "tokens_out": 2766, "ctx_chars": 0, "t_end": 1791535069.3895032, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.965, "lo95": 0.957, "mean": 0.9604, "ev_only": 0.9672, "ref_ms": 0.14739, "cand_ms": 0.15274, "n_trials": 215, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 71.52, "compile_s": 16.6, "r_error": "", "src_len": 2557, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 1440, "ts": 1791534892.189567, "chk": "ok", "code_len": 2557}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3198, "tout": 1326, "ts": 1791534997.519443, "chk": null, "code_len": 2557}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791534870.3341463}
{"ep": "h2-00441", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.7243, "speedup": 2.7243, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 229.1, "calls": 3, "tokens_in": 26598, "tokens_out": 14085, "ctx_chars": 8224, "t_end": 1791535070.6812353, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 2.7243, "lo95": 2.7182, "mean": 2.7855, "ev_only": 2.7258, "ref_ms": 59.98419, "cand_ms": 22.01829, "n_trials": 26, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.887, "r_wall_s": 26.45, "compile_s": 10.1, "r_error": "", "src_len": 5875, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 8214, "ts": 1791534913.7698047, "chk": "no_custom_kernel", "code_len": 5897}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9040, "tout": 3016, "ts": 1791534963.2416897, "chk": "ok", "code_len": 5875}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 12028, "tout": 2855, "ts": 1791535044.1081438, "chk": null, "code_len": 5875}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791534842.146784}
{"ep": "h2-00479", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8766, "speedup": 0.8766, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.3, "calls": 3, "tokens_in": 18899, "tokens_out": 1656, "ctx_chars": 8224, "t_end": 1791535077.5323188, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8766, "lo95": 0.6935, "mean": 0.7895, "ev_only": 0.8769, "ref_ms": 49.89421, "cand_ms": 56.91478, "n_trials": 25, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 63.11, "compile_s": 4.71, "r_error": "", "src_len": 1178, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 1043, "ts": 1791534965.2484503, "chk": "ok", "code_len": 1178}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6660, "tout": 47, "ts": 1791535008.7187989, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6731, "tout": 566, "ts": 1791535014.154279, "chk": null, "code_len": 1178}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791534952.2602072}
{"ep": "h2-00464", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0236, "speedup": 1.0236, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 173.9, "calls": 2, "tokens_in": 6315, "tokens_out": 7010, "ctx_chars": 0, "t_end": 1791535079.0160208, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0236, "lo95": 1.0228, "mean": 1.0233, "ev_only": 1.0238, "ref_ms": 30.50598, "cand_ms": 29.80181, "n_trials": 54, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0162, "r_wall_s": 95.19, "compile_s": 2.92, "r_error": "", "src_len": 4890, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 4473, "ts": 1791534944.414612, "chk": "ok", "code_len": 4890}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4690, "tout": 2537, "ts": 1791534982.8160625, "chk": null, "code_len": 4890}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791534905.6980512}
{"ep": "h2-00439", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.3148, "speedup": 1.3148, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 263.1, "calls": 3, "tokens_in": 10333, "tokens_out": 6935, "ctx_chars": 0, "t_end": 1791535083.6358476, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.3148, "lo95": 1.3084, "mean": 1.3138, "ev_only": 1.3204, "ref_ms": 15.51379, "cand_ms": 11.79953, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9767, "r_wall_s": 145.06, "compile_s": 3.46, "r_error": "", "src_len": 2338, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 2857, "ts": 1791534849.594598, "chk": "runtime_error", "code_len": 2369}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 3532, "tout": 2429, "ts": 1791534883.9074292, "chk": "ok", "code_len": 2338}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5234, "tout": 1649, "ts": 1791534938.2547717, "chk": null, "code_len": 2338}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["runtime_error", "ok", null], "first_tin": 1567, "started": 1791534820.4999228}
{"ep": "h2-00467", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9767, "speedup": 0.9767, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 170.4, "calls": 3, "tokens_in": 23797, "tokens_out": 2130, "ctx_chars": 11539, "t_end": 1791535084.9572966, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9767, "lo95": 0.9765, "mean": 0.9766, "ev_only": 0.9764, "ref_ms": 31.39525, "cand_ms": 32.14319, "n_trials": 47, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0185, "r_wall_s": 125.57, "compile_s": 2.12, "r_error": "", "src_len": 1906, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1181, "ts": 1791534926.6815815, "chk": "ok", "code_len": 1906}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8322, "tout": 97, "ts": 1791534951.2590902, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8443, "tout": 852, "ts": 1791534958.577424, "chk": null, "code_len": 1906}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7032, "started": 1791534914.51057}
{"ep": "h2-00497", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9175, "speedup": 0.9175, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.9, "calls": 2, "tokens_in": 14577, "tokens_out": 8831, "ctx_chars": 6513, "t_end": 1791535088.4721217, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.9175, "lo95": 0.9165, "mean": 0.9173, "ev_only": 0.9186, "ref_ms": 3.36672, "cand_ms": 3.66935, "n_trials": 194, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0963, "r_wall_s": 14.3, "compile_s": 1.86, "r_error": "", "src_len": 6726, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 5594, "ts": 1791535042.1734874, "chk": "ok", "code_len": 6726}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9129, "tout": 3237, "ts": 1791535074.1511338, "chk": null, "code_len": 6726}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791534995.6104486}
{"ep": "h2-00463", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0588, "speedup": 1.0588, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 197.5, "calls": 2, "tokens_in": 10501, "tokens_out": 2295, "ctx_chars": 6513, "t_end": 1791535090.1006124, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0588, "lo95": 1.0569, "mean": 1.0593, "ev_only": 1.0582, "ref_ms": 15.47976, "cand_ms": 14.62009, "n_trials": 91, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1324, "r_wall_s": 143.85, "compile_s": 2.16, "r_error": "", "src_len": 2402, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1179, "ts": 1791534902.995958, "chk": "ok", "code_len": 2402}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5896, "tout": 1116, "ts": 1791534945.5931385, "chk": null, "code_len": 2402}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791534892.5847127}
{"ep": "h2-00483", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9241, "speedup": 0.9241, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 123.1, "calls": 3, "tokens_in": 24261, "tokens_out": 2336, "ctx_chars": 11539, "t_end": 1791535091.359124, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9241, "lo95": 0.9235, "mean": 0.9555, "ev_only": 0.9249, "ref_ms": 7.80389, "cand_ms": 8.44523, "n_trials": 137, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0603, "r_wall_s": 50.37, "compile_s": 1.87, "r_error": "", "src_len": 1814, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1189, "ts": 1791534981.6423485, "chk": "ok", "code_len": 1814}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8411, "tout": 302, "ts": 1791535033.2041636, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8737, "tout": 845, "ts": 1791535040.319792, "chk": null, "code_len": 1814}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791534968.8591835}
{"ep": "h2-00506", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 5.2965, "speedup": 5.2965, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 87.5, "calls": 2, "tokens_in": 6428, "tokens_out": 6699, "ctx_chars": 0, "t_end": 1791535095.3562539, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 5.2965, "lo95": 5.2876, "mean": 5.2479, "ev_only": 5.4671, "ref_ms": 1.81088, "cand_ms": 0.34191, "n_trials": 185, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 19.5, "compile_s": 11.89, "r_error": "", "src_len": 4513, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4335, "ts": 1791535045.012936, "chk": "ok", "code_len": 4593}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4669, "tout": 2364, "ts": 1791535075.8078158, "chk": null, "code_len": 4513}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1759, "started": 1791535007.861852}
{"ep": "h2-00466", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9268, "speedup": 0.9268, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 197.3, "calls": 3, "tokens_in": 24291, "tokens_out": 2400, "ctx_chars": 11539, "t_end": 1791535107.255046, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9268, "lo95": 0.9233, "mean": 0.9214, "ev_only": 0.9258, "ref_ms": 7.87613, "cand_ms": 8.49795, "n_trials": 135, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0625, "r_wall_s": 124.05, "compile_s": 2.96, "r_error": "", "src_len": 1793, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1127, "ts": 1791534922.210044, "chk": "ok", "code_len": 1793}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8349, "tout": 456, "ts": 1791534974.633916, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8829, "tout": 817, "ts": 1791534982.2504694, "chk": null, "code_len": 1793}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791534909.9958224}
{"ep": "h2-00469", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0451, "speedup": 1.0451, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 174.9, "calls": 2, "tokens_in": 12994, "tokens_out": 2191, "ctx_chars": 9274, "t_end": 1791535107.8995123, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0451, "lo95": 1.0433, "mean": 1.045, "ev_only": 1.0426, "ref_ms": 8.33204, "cand_ms": 7.97216, "n_trials": 147, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0049, "r_wall_s": 125.28, "compile_s": 3.25, "r_error": "", "src_len": 1668, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1298, "ts": 1791534944.0203652, "chk": "ok", "code_len": 1668}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7202, "tout": 893, "ts": 1791534982.331555, "chk": null, "code_len": 1668}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791534932.9500406}
{"ep": "h2-00490", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.6842, "speedup": 0.6842, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 128.1, "calls": 2, "tokens_in": 14118, "tokens_out": 1658, "ctx_chars": 10850, "t_end": 1791535109.1218405, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.6842, "lo95": 0.6803, "mean": 0.7936, "ev_only": 0.7277, "ref_ms": 0.14506, "cand_ms": 0.212, "n_trials": 193, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 16.64, "compile_s": 1.91, "r_error": "", "src_len": 914, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1132, "ts": 1791534991.8517885, "chk": "ok", "code_len": 914}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7494, "tout": 526, "ts": 1791535092.2824097, "chk": null, "code_len": 914}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791534980.9745975}
{"ep": "h2-00522", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0191, "speedup": 1.0191, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 33.5, "calls": 1, "tokens_in": 1750, "tokens_out": 1679, "ctx_chars": 0, "t_end": 1791535111.0766208, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0191, "lo95": 1.0188, "mean": 1.0243, "ev_only": 1.0195, "ref_ms": 8.13996, "cand_ms": 7.9876, "n_trials": 142, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.011, "r_wall_s": 17.58, "compile_s": 2.47, "r_error": "", "src_len": 1445, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1750, "tout": 1679, "ts": 1791535093.4275553, "chk": null, "code_len": 1445}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1750, "started": 1791535077.5326169}
{"ep": "h2-00383", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.0702, "speedup": 0.0702, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 488.2, "calls": 4, "tokens_in": 51417, "tokens_out": 17428, "ctx_chars": 8224, "t_end": 1791535116.0403461, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.0702, "lo95": 0.0697, "mean": 0.0702, "ev_only": 0.0701, "ref_ms": 32.34855, "cand_ms": 461.10447, "n_trials": 10, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.9977, "r_wall_s": 118.31, "compile_s": 3.07, "r_error": "", "src_len": 3325, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 5593, "ts": 1791534693.2765594, "chk": "forbidden", "code_len": 3218}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 11262, "tout": 1634, "ts": 1791534708.8749423, "chk": "ok", "code_len": 3062}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 13008, "tout": 8519, "ts": 1791534971.094603, "chk": "ok", "code_len": 3325}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 21639, "tout": 1682, "ts": 1791534997.4780512, "chk": null, "code_len": 3325}], "n_turns": 4, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", "ok", null], "first_tin": 5508, "started": 1791534627.8460622}
{"ep": "h2-00524", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8028, "speedup": 0.8028, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 38.1, "calls": 3, "tokens_in": 18432, "tokens_out": 1386, "ctx_chars": 8224, "t_end": 1791535121.7571075, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8028, "lo95": 0.8015, "mean": 0.8038, "ev_only": 0.8027, "ref_ms": 33.10945, "cand_ms": 41.24128, "n_trials": 36, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 13.89, "compile_s": 2.02, "r_error": "", "src_len": 1056, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 795, "ts": 1791535092.9117436, "chk": "ok", "code_len": 1056}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6412, "tout": 76, "ts": 1791535102.855997, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6512, "tout": 515, "ts": 1791535107.8261833, "chk": null, "code_len": 1056}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791535083.6362002}
{"ep": "h2-00492", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9803, "speedup": 0.9803, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 139.9, "calls": 2, "tokens_in": 12172, "tokens_out": 2802, "ctx_chars": 9033, "t_end": 1791535122.768575, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9803, "lo95": 0.9792, "mean": 0.9813, "ev_only": 0.9796, "ref_ms": 15.53394, "cand_ms": 15.84597, "n_trials": 80, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0078, "r_wall_s": 85.89, "compile_s": 6.81, "r_error": "", "src_len": 1187, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 2056, "ts": 1791535002.086519, "chk": "ok", "code_len": 1187}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6682, "tout": 746, "ts": 1791535036.3479035, "chk": null, "code_len": 1187}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791534982.8680327}
{"ep": "h2-00516", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0059, "speedup": 1.0059, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.9, "calls": 3, "tokens_in": 20191, "tokens_out": 2830, "ctx_chars": 9266, "t_end": 1791535130.432178, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0059, "lo95": 1.0053, "mean": 1.0064, "ev_only": 1.0063, "ref_ms": 30.42907, "cand_ms": 30.24911, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0388, "r_wall_s": 45.91, "compile_s": 1.91, "r_error": "", "src_len": 2563, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1519, "ts": 1791535059.7223134, "chk": "ok", "code_len": 2563}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7245, "tout": 63, "ts": 1791535072.3944125, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7332, "tout": 1248, "ts": 1791535084.1128738, "chk": null, "code_len": 2563}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791535037.5547843}
{"ep": "h2-00530", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.843, "speedup": 0.843, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 29.2, "calls": 1, "tokens_in": 7030, "tokens_out": 1241, "ctx_chars": 11175, "t_end": 1791535136.4057276, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.843, "lo95": 0.8421, "mean": 0.8426, "ev_only": 0.8442, "ref_ms": 14.23295, "cand_ms": 16.88405, "n_trials": 65, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 17.09, "compile_s": 1.88, "r_error": "", "src_len": 2603, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1241, "ts": 1791535119.2019162, "chk": null, "code_len": 2603}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791535107.854602}
{"ep": "h2-00501", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4473, "speedup": 1.4473, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 149.6, "calls": 2, "tokens_in": 6382, "tokens_out": 7142, "ctx_chars": 0, "t_end": 1791535150.917485, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.4473, "lo95": 1.4459, "mean": 1.4223, "ev_only": 1.4481, "ref_ms": 15.47401, "cand_ms": 10.69134, "n_trials": 100, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2385, "r_wall_s": 68.44, "compile_s": 2.29, "r_error": "", "src_len": 4383, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1904, "tout": 5068, "ts": 1791535050.2249405, "chk": "ok", "code_len": 4383}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4478, "tout": 2074, "ts": 1791535080.9775977, "chk": null, "code_len": 4383}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1904, "started": 1791535001.2742078}
{"ep": "h2-00525", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0429, "speedup": 1.0429, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 68.4, "calls": 2, "tokens_in": 5087, "tokens_out": 3188, "ctx_chars": 0, "t_end": 1791535153.3153641, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0429, "lo95": 1.0422, "mean": 1.0943, "ev_only": 1.0437, "ref_ms": 8.30198, "cand_ms": 7.96073, "n_trials": 100, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0026, "r_wall_s": 23.02, "compile_s": 1.9, "r_error": "", "src_len": 1917, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 2180, "ts": 1791535105.212742, "chk": "ok", "code_len": 1957}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3321, "tout": 1008, "ts": 1791535130.2022889, "chk": null, "code_len": 1917}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791535084.9575808}
{"ep": "h2-00535", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0024, "speedup": 1.0024, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 34.2, "calls": 1, "tokens_in": 1501, "tokens_out": 699, "ctx_chars": 0, "t_end": 1791535155.973448, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0024, "lo95": 0.9986, "mean": 1.0055, "ev_only": 1.0053, "ref_ms": 8.04222, "cand_ms": 8.02272, "n_trials": 96, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0076, "r_wall_s": 26.96, "compile_s": 2.0, "r_error": "", "src_len": 1178, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1501, "tout": 699, "ts": 1791535128.9238455, "chk": null, "code_len": 1178}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1501, "started": 1791535121.7573268}
{"ep": "h2-00527", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0205, "speedup": 1.0205, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 69.7, "calls": 2, "tokens_in": 4078, "tokens_out": 1940, "ctx_chars": 0, "t_end": 1791535159.7599344, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0205, "lo95": 1.0164, "mean": 1.0168, "ev_only": 1.0182, "ref_ms": 8.0959, "cand_ms": 7.93353, "n_trials": 133, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0046, "r_wall_s": 38.33, "compile_s": 2.68, "r_error": "", "src_len": 1138, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1335, "ts": 1791535102.4934742, "chk": "ok", "code_len": 1138}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2477, "tout": 605, "ts": 1791535121.3574262, "chk": null, "code_len": 1138}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791535090.100927}
{"ep": "h2-00526", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 77.4, "calls": 1, "tokens_in": 2743, "tokens_out": 2627, "ctx_chars": 0, "t_end": 1791535165.9078584, "error": "RuntimeError: Error building extension 'resnet_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=resnet_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 53.54, "compile_s": 53.53, "r_error": "RuntimeError: Error building extension 'resnet_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=resnet_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "src_len": 5554, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2743, "tout": 2627, "ts": 1791535112.3451934, "chk": null, "code_len": 5554}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2743, "started": 1791535088.4724119}
{"ep": "h2-00519", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0289, "speedup": 1.0289, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.0, "calls": 1, "tokens_in": 5832, "tokens_out": 542, "ctx_chars": 9656, "t_end": 1791535168.0071757, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0289, "lo95": 1.0277, "mean": 1.0289, "ev_only": 1.0281, "ref_ms": 24.93347, "cand_ms": 24.23241, "n_trials": 60, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0223, "r_wall_s": 92.0, "compile_s": 2.06, "r_error": "", "src_len": 992, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 542, "ts": 1791535075.148808, "chk": null, "code_len": 992}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791535068.032544}
{"ep": "h2-00514", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.787, "speedup": 0.787, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 142.7, "calls": 3, "tokens_in": 50127, "tokens_out": 14996, "ctx_chars": 2425, "t_end": 1791535169.1493893, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.787, "lo95": 0.7839, "mean": 0.7901, "ev_only": 0.7872, "ref_ms": 18.8225, "cand_ms": 23.91679, "n_trials": 55, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.03, "r_wall_s": 14.44, "compile_s": 2.21, "r_error": "", "src_len": 17570, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7580, "ts": 1791535086.903196, "chk": "ok", "code_len": 17646}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19240, "tout": 75, "ts": 1791535097.998415, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19339, "tout": 7341, "ts": 1791535154.6444697, "chk": null, "code_len": 17570}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791535026.499158}
{"ep": "h2-00470", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.015, "speedup": 1.015, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 234.9, "calls": 3, "tokens_in": 24763, "tokens_out": 2246, "ctx_chars": 12352, "t_end": 1791535169.926421, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.015, "lo95": 1.0144, "mean": 1.0149, "ev_only": 1.0158, "ref_ms": 64.54961, "cand_ms": 63.59513, "n_trials": 27, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 166.03, "compile_s": 3.15, "r_error": "", "src_len": 1912, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1259, "ts": 1791534947.379152, "chk": "ok", "code_len": 1912}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8699, "tout": 13, "ts": 1791534995.3053513, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8736, "tout": 974, "ts": 1791535003.62706, "chk": null, "code_len": 1912}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791534935.0016596}
{"ep": "h2-00529", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0427, "speedup": 1.0427, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.3, "calls": 2, "tokens_in": 13019, "tokens_out": 2278, "ctx_chars": 9274, "t_end": 1791535172.7037854, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0427, "lo95": 1.042, "mean": 1.0487, "ev_only": 1.0437, "ref_ms": 8.39174, "cand_ms": 8.04793, "n_trials": 157, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0054, "r_wall_s": 43.65, "compile_s": 1.97, "r_error": "", "src_len": 1785, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1323, "ts": 1791535107.0022247, "chk": "ok", "code_len": 1785}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7227, "tout": 955, "ts": 1791535128.9154453, "chk": null, "code_len": 1785}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791535095.3565848}
{"ep": "h2-00380", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4965, "speedup": 1.4965, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 553.0, "calls": 3, "tokens_in": 15845, "tokens_out": 12276, "ctx_chars": 0, "t_end": 1791535173.4133482, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.4965, "lo95": 1.4877, "mean": 1.4888, "ev_only": 1.4953, "ref_ms": 7.86722, "cand_ms": 5.25712, "n_trials": 182, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 123.8, "compile_s": 11.94, "r_error": "", "src_len": 5056, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6567, "ts": 1791534679.5649276, "chk": "no_custom_kernel", "code_len": 5060}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5455, "tout": 3113, "ts": 1791534932.1918209, "chk": "ok", "code_len": 5056}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8231, "tout": 2596, "ts": 1791535048.9432747, "chk": null, "code_len": 5056}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791534620.4059923}
{"ep": "h2-00517", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4819, "speedup": 1.4819, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 130.1, "calls": 2, "tokens_in": 11101, "tokens_out": 5105, "ctx_chars": 6513, "t_end": 1791535187.665648, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.4819, "lo95": 1.4809, "mean": 1.4784, "ev_only": 1.4805, "ref_ms": 15.49378, "cand_ms": 10.45501, "n_trials": 93, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1916, "r_wall_s": 70.54, "compile_s": 2.4, "r_error": "", "src_len": 3239, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3641, "ts": 1791535092.365536, "chk": "ok", "code_len": 3239}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6494, "tout": 1464, "ts": 1791535116.280129, "chk": null, "code_len": 3239}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791535057.5896614}
{"ep": "h2-00484", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0095, "speedup": 1.0095, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 219.6, "calls": 2, "tokens_in": 12861, "tokens_out": 2802, "ctx_chars": 9266, "t_end": 1791535189.090729, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0095, "lo95": 1.0063, "mean": 1.0079, "ev_only": 1.0079, "ref_ms": 30.64169, "cand_ms": 30.35266, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0388, "r_wall_s": 82.26, "compile_s": 1.99, "r_error": "", "src_len": 2434, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1521, "ts": 1791534983.4150977, "chk": "ok", "code_len": 2434}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7247, "tout": 1281, "ts": 1791535106.4005668, "chk": null, "code_len": 2434}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5614, "started": 1791534969.4952924}
{"ep": "h2-00480", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.6673, "speedup": 1.6673, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 239.5, "calls": 2, "tokens_in": 30339, "tokens_out": 21696, "ctx_chars": 0, "t_end": 1791535193.3898273, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.6673, "lo95": 1.4747, "mean": 1.653, "ev_only": 1.6754, "ref_ms": 19.16433, "cand_ms": 11.49393, "n_trials": 50, "fast1": true, "flags": [], "backend": "mixed", "own_frac": 0.1605, "r_wall_s": 49.94, "compile_s": 13.09, "r_error": "", "src_len": 18838, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 13269, "ts": 1791535067.4998093, "chk": "ok", "code_len": 18838}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19705, "tout": 8427, "ts": 1791535143.4113436, "chk": null, "code_len": 18838}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791534953.859926}
{"ep": "h2-00537", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.837, "speedup": 0.837, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 64.6, "calls": 3, "tokens_in": 21124, "tokens_out": 958, "ctx_chars": 10850, "t_end": 1791535195.0376914, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.837, "lo95": 0.8323, "mean": 0.8137, "ev_only": 0.7935, "ref_ms": 0.15403, "cand_ms": 0.18403, "n_trials": 174, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 35.51, "compile_s": 2.07, "r_error": "", "src_len": 834, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 469, "ts": 1791535137.3371382, "chk": "ok", "code_len": 834}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7200, "tout": 78, "ts": 1791535155.2123423, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7302, "tout": 411, "ts": 1791535159.3702846, "chk": null, "code_len": 834}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6622, "started": 1791535130.4324303}
{"ep": "h2-00520", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4385, "speedup": 1.4385, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 130.7, "calls": 2, "tokens_in": 11165, "tokens_out": 5161, "ctx_chars": 6513, "t_end": 1791535200.0913067, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.4385, "lo95": 1.4357, "mean": 1.4365, "ev_only": 1.4394, "ref_ms": 15.59057, "cand_ms": 10.83785, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1641, "r_wall_s": 69.03, "compile_s": 2.09, "r_error": "", "src_len": 3234, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3652, "ts": 1791535103.4920368, "chk": "ok", "code_len": 3234}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6558, "tout": 1509, "ts": 1791535130.3019738, "chk": null, "code_len": 3234}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791535069.3897865}
{"ep": "h2-00503", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.6464, "speedup": 2.6464, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 200.6, "calls": 3, "tokens_in": 27141, "tokens_out": 17421, "ctx_chars": 8224, "t_end": 1791535203.4613836, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.6464, "lo95": 2.6418, "mean": 2.6461, "ev_only": 2.6476, "ref_ms": 60.13212, "cand_ms": 22.72228, "n_trials": 25, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8917, "r_wall_s": 20.53, "compile_s": 11.04, "r_error": "", "src_len": 6379, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 10588, "ts": 1791535097.9178724, "chk": "no_custom_kernel", "code_len": 6562}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9182, "tout": 3781, "ts": 1791535145.878405, "chk": "ok", "code_len": 6379}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 12429, "tout": 3052, "ts": 1791535182.8705554, "chk": null, "code_len": 6379}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791535002.8664255}
{"ep": "h2-00476", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0301, "speedup": 1.0301, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 260.2, "calls": 2, "tokens_in": 4493, "tokens_out": 2361, "ctx_chars": 0, "t_end": 1791535205.5586917, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0301, "lo95": 1.0292, "mean": 1.0297, "ev_only": 1.029, "ref_ms": 24.92717, "cand_ms": 24.19978, "n_trials": 32, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0289, "r_wall_s": 185.33, "compile_s": 7.11, "r_error": "", "src_len": 1054, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1760, "ts": 1791534960.3636477, "chk": "ok", "code_len": 1054}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2825, "tout": 601, "ts": 1791535018.5885873, "chk": null, "code_len": 1054}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791534945.3212087}
{"ep": "h2-00545", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.003, "speedup": 1.003, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 41.8, "calls": 2, "tokens_in": 11396, "tokens_out": 1094, "ctx_chars": 9028, "t_end": 1791535210.9768634, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.003, "lo95": 1.0026, "mean": 1.002, "ev_only": 1.0032, "ref_ms": 8.0323, "cand_ms": 8.00842, "n_trials": 171, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0083, "r_wall_s": 23.01, "compile_s": 1.87, "r_error": "", "src_len": 1024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 546, "ts": 1791535174.5506952, "chk": "ok", "code_len": 1024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6027, "tout": 548, "ts": 1791535187.8925762, "chk": null, "code_len": 1024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791535169.149659}
{"ep": "h2-00402", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4645, "speedup": 1.4645, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 552.8, "calls": 2, "tokens_in": 11228, "tokens_out": 5338, "ctx_chars": 6513, "t_end": 1791535213.1402438, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.4645, "lo95": 1.4585, "mean": 1.4532, "ev_only": 1.4653, "ref_ms": 15.55814, "cand_ms": 10.62365, "n_trials": 42, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1838, "r_wall_s": 214.98, "compile_s": 2.89, "r_error": "", "src_len": 3285, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3810, "ts": 1791534697.3693035, "chk": "ok", "code_len": 3285}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6621, "tout": 1528, "ts": 1791534997.2530437, "chk": null, "code_len": 3285}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791534660.365781}
{"ep": "h2-00285", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8058, "speedup": 0.8058, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 937.2, "calls": 5, "tokens_in": 60562, "tokens_out": 14443, "ctx_chars": 6513, "t_end": 1791535225.2474113, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8058, "lo95": 0.7973, "mean": 0.7312, "ev_only": 0.8035, "ref_ms": 7.48066, "cand_ms": 9.28362, "n_trials": 132, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.087, "r_wall_s": 238.8, "compile_s": 3.01, "r_error": "", "src_len": 5060, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2819, "ts": 1791534311.8270028, "chk": "incorrect", "code_len": 6316}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8596, "tout": 3639, "ts": 1791534370.9109507, "chk": "incorrect", "code_len": 5991}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 12367, "tout": 3003, "ts": 1791534666.8696265, "chk": "incorrect", "code_len": 4966}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 15502, "tout": 2818, "ts": 1791534951.279118, "chk": "ok", "code_len": 5060}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 18452, "tout": 2164, "ts": 1791534986.198684, "chk": null, "code_len": 5060}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "incorrect", "ok", null], "first_tin": 5645, "started": 1791534288.0249975}
{"ep": "h2-00542", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.5979, "speedup": 0.5979, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 71.9, "calls": 1, "tokens_in": 6622, "tokens_out": 496, "ctx_chars": 10850, "t_end": 1791535231.620259, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.5979, "lo95": 0.5773, "mean": 0.6701, "ev_only": 0.5902, "ref_ms": 0.16432, "cand_ms": 0.27484, "n_trials": 227, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 66.29, "compile_s": 2.25, "r_error": "", "src_len": 834, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6622, "tout": 496, "ts": 1791535165.1546721, "chk": null, "code_len": 834}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6622, "started": 1791535159.7601626}
{"ep": "h2-00502", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0195, "speedup": 1.0195, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 230.4, "calls": 4, "tokens_in": 35584, "tokens_out": 4248, "ctx_chars": 11823, "t_end": 1791535231.7642817, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0195, "lo95": 1.0189, "mean": 1.0197, "ev_only": 1.0195, "ref_ms": 27.51251, "cand_ms": 26.98566, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0147, "r_wall_s": 13.75, "compile_s": 1.98, "r_error": "", "src_len": 2430, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1508, "ts": 1791535016.3322792, "chk": "ok", "code_len": 2676}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8541, "tout": 1378, "ts": 1791535194.1364915, "chk": "ok", "code_len": 2430}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10031, "tout": 36, "ts": 1791535205.613984, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 10091, "tout": 1326, "ts": 1791535217.971236, "chk": null, "code_len": 2430}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 6921, "started": 1791535001.3397048}
{"ep": "h2-00544", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.6557, "speedup": 2.6557, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 65.3, "calls": 2, "tokens_in": 13094, "tokens_out": 4418, "ctx_chars": 8529, "t_end": 1791535233.3072999, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.6557, "lo95": 2.6534, "mean": 2.6515, "ev_only": 2.6636, "ref_ms": 8.50622, "cand_ms": 3.20296, "n_trials": 67, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3433, "r_wall_s": 15.9, "compile_s": 2.03, "r_error": "", "src_len": 2943, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2789, "ts": 1791535192.4282947, "chk": "ok", "code_len": 2943}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7597, "tout": 1629, "ts": 1791535217.3487883, "chk": null, "code_len": 2943}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791535168.5578957}
{"ep": "h2-00523", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9781, "speedup": 0.9781, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 159.3, "calls": 2, "tokens_in": 12317, "tokens_out": 2978, "ctx_chars": 9033, "t_end": 1791535238.3210697, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9781, "lo95": 0.969, "mean": 0.9741, "ev_only": 0.9787, "ref_ms": 15.54881, "cand_ms": 15.89688, "n_trials": 71, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0127, "r_wall_s": 79.39, "compile_s": 2.09, "r_error": "", "src_len": 1242, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 2125, "ts": 1791535104.454803, "chk": "ok", "code_len": 1242}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6827, "tout": 853, "ts": 1791535158.444061, "chk": null, "code_len": 1242}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791535079.0163343}
{"ep": "h2-00534", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.2711, "speedup": 0.2711, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 127.4, "calls": 4, "tokens_in": 36961, "tokens_out": 7973, "ctx_chars": 6513, "t_end": 1791535243.4322507, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.2711, "lo95": 0.2689, "mean": 0.2488, "ev_only": 0.2707, "ref_ms": 7.191, "cand_ms": 26.52769, "n_trials": 55, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0526, "r_wall_s": 37.6, "compile_s": 2.02, "r_error": "", "src_len": 5603, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2716, "ts": 1791535139.9748526, "chk": "incorrect", "code_len": 6148}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8493, "tout": 2772, "ts": 1791535172.568193, "chk": "ok", "code_len": 5603}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11377, "tout": 45, "ts": 1791535183.9043121, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11446, "tout": 2440, "ts": 1791535205.8051443, "chk": null, "code_len": 5603}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791535116.0406058}
{"ep": "h2-00462", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0032, "speedup": 1.0032, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 352.1, "calls": 3, "tokens_in": 19801, "tokens_out": 2104, "ctx_chars": 9274, "t_end": 1791535244.54398, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0032, "lo95": 1.0004, "mean": 1.012, "ev_only": 1.0062, "ref_ms": 8.41911, "cand_ms": 8.39216, "n_trials": 121, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0422, "r_wall_s": 241.35, "compile_s": 3.24, "r_error": "", "src_len": 2021, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1085, "ts": 1791534903.0791032, "chk": "ok", "code_len": 2021}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6987, "tout": 13, "ts": 1791534994.485773, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7024, "tout": 1006, "ts": 1791535002.8947756, "chk": null, "code_len": 2021}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791534892.4769068}
{"ep": "h2-00473", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.9083, "speedup": 3.9083, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 304.5, "calls": 2, "tokens_in": 7623, "tokens_out": 11442, "ctx_chars": 0, "t_end": 1791535246.5106828, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 3.9083, "lo95": 3.8561, "mean": 3.8947, "ev_only": 3.9459, "ref_ms": 8.63462, "cand_ms": 2.20931, "n_trials": 151, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9973, "r_wall_s": 44.7, "compile_s": 12.39, "r_error": "", "src_len": 7541, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 7906, "ts": 1791535012.6430442, "chk": "ok", "code_len": 7561}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5803, "tout": 3536, "ts": 1791535201.7709022, "chk": null, "code_len": 7541}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791534941.997497}
{"ep": "h2-00510", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0544, "speedup": 1.0544, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 232.3, "calls": 4, "tokens_in": 19356, "tokens_out": 4346, "ctx_chars": 0, "t_end": 1791535249.8273427, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0544, "lo95": 1.0509, "mean": 1.0533, "ev_only": 1.0546, "ref_ms": 19.65425, "cand_ms": 18.64011, "n_trials": 58, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0085, "r_wall_s": 59.41, "compile_s": 11.43, "r_error": "", "src_len": 2864, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1956, "tout": 1619, "ts": 1791535032.8852696, "chk": "compile_error", "code_len": 3117}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4803, "tout": 1342, "ts": 1791535114.105809, "chk": "ok", "code_len": 2864}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6257, "tout": 59, "ts": 1791535179.3571904, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6340, "tout": 1326, "ts": 1791535190.3261847, "chk": null, "code_len": 2864}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 1956, "started": 1791535017.5540085}
{"ep": "h2-00512", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9755, "speedup": 0.9755, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 230.2, "calls": 3, "tokens_in": 24026, "tokens_out": 2354, "ctx_chars": 11539, "t_end": 1791535250.191556, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9755, "lo95": 0.9753, "mean": 0.9755, "ev_only": 0.9759, "ref_ms": 31.49354, "cand_ms": 32.28586, "n_trials": 40, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0184, "r_wall_s": 89.72, "compile_s": 2.0, "r_error": "", "src_len": 1830, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1169, "ts": 1791535031.8586063, "chk": "ok", "code_len": 1830}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8310, "tout": 350, "ts": 1791535152.0602634, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8684, "tout": 835, "ts": 1791535159.6253555, "chk": null, "code_len": 1830}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7032, "started": 1791535020.032744}
{"ep": "h2-00536", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8957, "speedup": 0.8957, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 129.2, "calls": 4, "tokens_in": 29571, "tokens_out": 1476, "ctx_chars": 10850, "t_end": 1791535251.9969585, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8957, "lo95": 0.8118, "mean": 0.8317, "ev_only": 0.8701, "ref_ms": 0.17505, "cand_ms": 0.19544, "n_trials": 184, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 85.22, "compile_s": 3.28, "r_error": "", "src_len": 837, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 498, "ts": 1791535128.048039, "chk": "ok", "code_len": 856}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 7229, "tout": 476, "ts": 1791535151.0058699, "chk": "ok", "code_len": 837}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7814, "tout": 68, "ts": 1791535162.28732, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7906, "tout": 434, "ts": 1791535166.6335196, "chk": null, "code_len": 837}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 6622, "started": 1791535122.7688608}
{"ep": "h2-00546", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0094, "speedup": 1.0094, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 82.1, "calls": 2, "tokens_in": 4006, "tokens_out": 2071, "ctx_chars": 0, "t_end": 1791535252.042514, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0094, "lo95": 1.008, "mean": 1.0091, "ev_only": 1.0076, "ref_ms": 8.02609, "cand_ms": 7.95162, "n_trials": 187, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 54.85, "compile_s": 1.93, "r_error": "", "src_len": 910, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1484, "ts": 1791535183.5140636, "chk": "ok", "code_len": 910}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2503, "tout": 587, "ts": 1791535197.104372, "chk": null, "code_len": 910}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791535169.9266622}
{"ep": "h2-00551", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9819, "speedup": 0.9819, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 62.7, "calls": 1, "tokens_in": 5488, "tokens_out": 752, "ctx_chars": 9033, "t_end": 1791535256.0669537, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9819, "lo95": 0.9817, "mean": 0.9989, "ev_only": 0.982, "ref_ms": 15.47458, "cand_ms": 15.75929, "n_trials": 98, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0177, "r_wall_s": 53.16, "compile_s": 1.99, "r_error": "", "src_len": 718, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5488, "tout": 752, "ts": 1791535202.599998, "chk": null, "code_len": 718}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5488, "started": 1791535193.390114}
{"ep": "h2-00552", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0635, "speedup": 1.0635, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 63.3, "calls": 2, "tokens_in": 4965, "tokens_out": 3585, "ctx_chars": 0, "t_end": 1791535258.3644316, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0635, "lo95": 1.063, "mean": 1.0825, "ev_only": 1.0636, "ref_ms": 8.47564, "cand_ms": 7.96964, "n_trials": 136, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0054, "r_wall_s": 24.3, "compile_s": 1.88, "r_error": "", "src_len": 2230, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 2352, "ts": 1791535215.3796031, "chk": "ok", "code_len": 2230}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3330, "tout": 1233, "ts": 1791535233.993988, "chk": null, "code_len": 2230}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791535195.0381181}
{"ep": "h2-00509", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3348, "speedup": 1.3348, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 242.3, "calls": 1, "tokens_in": 5998, "tokens_out": 723, "ctx_chars": 9970, "t_end": 1791535259.0429235, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.3348, "lo95": 1.3342, "mean": 1.3343, "ev_only": 1.3324, "ref_ms": 7.82006, "cand_ms": 5.85868, "n_trials": 190, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1106, "r_wall_s": 234.35, "compile_s": 7.71, "r_error": "", "src_len": 1452, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 723, "ts": 1791535023.3613064, "chk": null, "code_len": 1452}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791535016.6997435}
{"ep": "h2-00549", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0134, "speedup": 1.0134, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 74.4, "calls": 2, "tokens_in": 10546, "tokens_out": 3239, "ctx_chars": 5026, "t_end": 1791535262.0722847, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0134, "lo95": 1.0118, "mean": 1.0159, "ev_only": 1.0165, "ref_ms": 8.15376, "cand_ms": 8.04578, "n_trials": 142, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.009, "r_wall_s": 37.77, "compile_s": 1.99, "r_error": "", "src_len": 1455, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 2448, "ts": 1791535209.1030161, "chk": "ok", "code_len": 1455}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6553, "tout": 791, "ts": 1791535224.1515753, "chk": null, "code_len": 1455}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791535187.6659184}
{"ep": "h2-00547", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.6744, "speedup": 2.6744, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.6, "calls": 2, "tokens_in": 12934, "tokens_out": 4145, "ctx_chars": 8529, "t_end": 1791535263.2594864, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 2.6744, "lo95": 2.6732, "mean": 2.6724, "ev_only": 2.6708, "ref_ms": 8.48718, "cand_ms": 3.17354, "n_trials": 197, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3408, "r_wall_s": 49.11, "compile_s": 2.64, "r_error": "", "src_len": 2874, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2580, "ts": 1791535195.0955536, "chk": "ok", "code_len": 2874}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7437, "tout": 1565, "ts": 1791535214.1055121, "chk": null, "code_len": 2874}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791535172.7040305}
{"ep": "h2-00528", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 37.0805, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 173.7, "calls": 2, "tokens_in": 7100, "tokens_out": 7101, "ctx_chars": 0, "t_end": 1791535265.0125403, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 37.0805, "lo95": 37.0165, "mean": 36.3951, "ev_only": 37.065, "ref_ms": 19.61413, "cand_ms": 0.52896, "n_trials": 68, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 70.25, "compile_s": 13.1, "r_error": "", "src_len": 5022, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4616, "ts": 1791535131.156815, "chk": "ok", "code_len": 5022}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5142, "tout": 2485, "ts": 1791535194.7270179, "chk": null, "code_len": 5022}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791535091.3593943}
{"ep": "h2-00482", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0178, "speedup": 1.0178, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 308.7, "calls": 2, "tokens_in": 15478, "tokens_out": 3303, "ctx_chars": 11823, "t_end": 1791535268.0451949, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0178, "lo95": 1.0169, "mean": 1.0178, "ev_only": 1.0176, "ref_ms": 27.53563, "cand_ms": 27.05357, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0118, "r_wall_s": 246.7, "compile_s": 5.72, "r_error": "", "src_len": 2214, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 2094, "ts": 1791534983.8548274, "chk": "ok", "code_len": 2214}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8555, "tout": 1209, "ts": 1791535021.0898736, "chk": null, "code_len": 2214}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791534959.331258}
{"ep": "h2-00515", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3478, "speedup": 1.3478, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 234.1, "calls": 2, "tokens_in": 14707, "tokens_out": 16453, "ctx_chars": 6513, "t_end": 1791535271.5262663, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.3478, "lo95": 1.3437, "mean": 1.3445, "ev_only": 1.3489, "ref_ms": 15.77988, "cand_ms": 11.70756, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9703, "r_wall_s": 53.33, "compile_s": 1.92, "r_error": "", "src_len": 8701, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 12110, "ts": 1791535143.4192448, "chk": "ok", "code_len": 8701}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9843, "tout": 4343, "ts": 1791535217.9474413, "chk": null, "code_len": 8701}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535038.0056472}
{"ep": "h2-00543", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0652, "speedup": 1.0652, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 105.7, "calls": 2, "tokens_in": 10492, "tokens_out": 2293, "ctx_chars": 6513, "t_end": 1791535271.6006653, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0652, "lo95": 1.0646, "mean": 1.0642, "ev_only": 1.0656, "ref_ms": 15.49932, "cand_ms": 14.5503, "n_trials": 92, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1156, "r_wall_s": 71.17, "compile_s": 1.99, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1170, "ts": 1791535176.0433915, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5887, "tout": 1123, "ts": 1791535199.734406, "chk": null, "code_len": 2422}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791535165.908125}
{"ep": "h2-00456", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.7077, "speedup": 0.7077, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 404.4, "calls": 2, "tokens_in": 5942, "tokens_out": 7912, "ctx_chars": 0, "t_end": 1791535273.6644564, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7077, "lo95": 0.7016, "mean": 0.7061, "ev_only": 0.7065, "ref_ms": 7.84955, "cand_ms": 11.09218, "n_trials": 104, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.9995, "r_wall_s": 242.85, "compile_s": 20.51, "r_error": "", "src_len": 4066, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 5929, "ts": 1791534927.5892982, "chk": "ok", "code_len": 4066}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4217, "tout": 1983, "ts": 1791535029.489279, "chk": null, "code_len": 4066}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1725, "started": 1791534869.224732}
{"ep": "h2-00471", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0104, "speedup": 1.0104, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 336.7, "calls": 2, "tokens_in": 4178, "tokens_out": 2230, "ctx_chars": 0, "t_end": 1791535274.9128532, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0104, "lo95": 1.0091, "mean": 1.0119, "ev_only": 1.0122, "ref_ms": 8.01582, "cand_ms": 7.9331, "n_trials": 136, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 272.44, "compile_s": 3.54, "r_error": "", "src_len": 1069, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1433, "ts": 1791534951.940526, "chk": "ok", "code_len": 1069}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2577, "tout": 797, "ts": 1791535002.1769054, "chk": null, "code_len": 1069}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791534938.207462}
{"ep": "h2-00554", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9756, "speedup": 0.9756, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 79.4, "calls": 1, "tokens_in": 7032, "tokens_out": 1339, "ctx_chars": 11539, "t_end": 1791535282.9072704, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9756, "lo95": 0.9755, "mean": 0.9755, "ev_only": 0.976, "ref_ms": 31.36451, "cand_ms": 32.14815, "n_trials": 48, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.02, "r_wall_s": 63.26, "compile_s": 1.99, "r_error": "", "src_len": 1976, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1339, "ts": 1791535218.805184, "chk": null, "code_len": 1976}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791535203.461607}
{"ep": "h2-00494", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.6177, "speedup": 0.6177, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 301.9, "calls": 3, "tokens_in": 18762, "tokens_out": 1537, "ctx_chars": 8224, "t_end": 1791535288.4645872, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.6177, "lo95": 0.519, "mean": 0.7612, "ev_only": 0.6172, "ref_ms": 32.6521, "cand_ms": 52.86465, "n_trials": 39, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0004, "r_wall_s": 15.82, "compile_s": 2.37, "r_error": "", "src_len": 1029, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 956, "ts": 1791534997.8842661, "chk": "ok", "code_len": 1029}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6573, "tout": 84, "ts": 1791535267.9451191, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6681, "tout": 497, "ts": 1791535272.6137471, "chk": null, "code_len": 1029}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791534986.6064856}
{"ep": "h2-00548", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 118.4, "calls": 1, "tokens_in": 1818, "tokens_out": 8671, "ctx_chars": 0, "t_end": 1791535291.8342736, "error": "RuntimeError: Error building extension 'fused_lse_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_lse_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 47.8, "compile_s": 47.79, "r_error": "RuntimeError: Error building extension 'fused_lse_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_lse_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "src_len": 3831, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1818, "tout": 8671, "ts": 1791535244.0161808, "chk": null, "code_len": 3831}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1818, "started": 1791535173.4136112}
{"ep": "h2-00532", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9488, "speedup": 0.9488, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 189.6, "calls": 2, "tokens_in": 13865, "tokens_out": 12659, "ctx_chars": 6513, "t_end": 1791535298.755108, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9488, "lo95": 0.9479, "mean": 0.9509, "ev_only": 0.9509, "ref_ms": 15.7328, "cand_ms": 16.58174, "n_trials": 85, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.5752, "r_wall_s": 70.46, "compile_s": 1.99, "r_error": "", "src_len": 6936, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 9215, "ts": 1791535190.2676346, "chk": "ok", "code_len": 6936}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9001, "tout": 3444, "ts": 1791535228.1192, "chk": null, "code_len": 6936}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535109.122115}
{"ep": "h2-00518", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.717, "speedup": 2.717, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 238.8, "calls": 2, "tokens_in": 7886, "tokens_out": 10400, "ctx_chars": 0, "t_end": 1791535306.0813184, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 2.717, "lo95": 2.7146, "mean": 2.7162, "ev_only": 2.7134, "ref_ms": 14.32478, "cand_ms": 5.2723, "n_trials": 102, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 106.87, "compile_s": 12.93, "r_error": "", "src_len": 5953, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 7542, "ts": 1791535133.4987285, "chk": "ok", "code_len": 5953}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5703, "tout": 2858, "ts": 1791535199.1193635, "chk": null, "code_len": 5953}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2183, "started": 1791535067.2370331}
{"ep": "h2-00385", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0624, "speedup": 1.0624, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 676.0, "calls": 2, "tokens_in": 4781, "tokens_out": 3090, "ctx_chars": 0, "t_end": 1791535309.4415274, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0624, "lo95": 1.057, "mean": 1.0598, "ev_only": 1.0657, "ref_ms": 8.51677, "cand_ms": 8.01677, "n_trials": 124, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0052, "r_wall_s": 324.88, "compile_s": 3.31, "r_error": "", "src_len": 2024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 1951, "ts": 1791534654.6993117, "chk": "ok", "code_len": 2024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3146, "tout": 1139, "ts": 1791534984.2622943, "chk": null, "code_len": 2024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791534633.3942788}
{"ep": "h2-00531", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0107, "speedup": 1.0107, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 205.2, "calls": 2, "tokens_in": 6287, "tokens_out": 7241, "ctx_chars": 0, "t_end": 1791535313.1466198, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0107, "lo95": 1.0103, "mean": 1.0103, "ev_only": 1.0112, "ref_ms": 30.10498, "cand_ms": 29.78576, "n_trials": 43, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0164, "r_wall_s": 53.76, "compile_s": 2.06, "r_error": "", "src_len": 4761, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 4727, "ts": 1791535147.929991, "chk": "ok", "code_len": 4761}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4689, "tout": 2514, "ts": 1791535258.9031932, "chk": null, "code_len": 4761}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791535107.8997412}
{"ep": "h2-00521", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.003, "speedup": 1.003, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 242.5, "calls": 5, "tokens_in": 25202, "tokens_out": 3949, "ctx_chars": 0, "t_end": 1791535313.1741154, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.003, "lo95": 1.0022, "mean": 1.0033, "ev_only": 1.0026, "ref_ms": 42.54376, "cand_ms": 42.41598, "n_trials": 35, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0, "r_wall_s": 83.7, "compile_s": 10.79, "r_error": "", "src_len": 1676, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2034, "tout": 1156, "ts": 1791535083.0572875, "chk": "compile_error", "code_len": 1935}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4413, "tout": 888, "ts": 1791535135.9628503, "chk": "ok", "code_len": 1969}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 5413, "tout": 1097, "ts": 1791535195.6907074, "chk": "ok", "code_len": 1676}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6622, "tout": 51, "ts": 1791535221.6836972, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 6720, "tout": 757, "ts": 1791535228.7742748, "chk": null, "code_len": 1676}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "ok", "no_code", null], "first_tin": 2034, "started": 1791535070.681474}
{"ep": "h2-00486", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0681, "speedup": 1.0681, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 343.3, "calls": 2, "tokens_in": 13853, "tokens_out": 3289, "ctx_chars": 10167, "t_end": 1791535314.1781096, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0681, "lo95": 1.0645, "mean": 1.0653, "ev_only": 1.0666, "ref_ms": 8.60559, "cand_ms": 8.05662, "n_trials": 152, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0053, "r_wall_s": 291.34, "compile_s": 7.15, "r_error": "", "src_len": 2159, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 2094, "ts": 1791534988.3259861, "chk": "ok", "code_len": 2159}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7747, "tout": 1195, "ts": 1791535022.4886878, "chk": null, "code_len": 2159}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791534970.8372393}
{"ep": "h2-00582", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8015, "speedup": 0.8015, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 33.8, "calls": 1, "tokens_in": 1984, "tokens_out": 736, "ctx_chars": 0, "t_end": 1791535322.2619615, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8015, "lo95": 0.7984, "mean": 0.8011, "ev_only": 0.8015, "ref_ms": 60.75066, "cand_ms": 75.79395, "n_trials": 14, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 25.32, "compile_s": 1.83, "r_error": "", "src_len": 939, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1984, "tout": 736, "ts": 1791535296.8838725, "chk": null, "code_len": 939}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1984, "started": 1791535288.4648705}
{"ep": "h2-00539", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.3204, "speedup": 2.3204, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 171.7, "calls": 3, "tokens_in": 16220, "tokens_out": 12499, "ctx_chars": 0, "t_end": 1791535322.5864847, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 2.3204, "lo95": 2.3185, "mean": 2.3202, "ev_only": 2.3246, "ref_ms": 14.30314, "cand_ms": 6.16417, "n_trials": 123, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9993, "r_wall_s": 37.29, "compile_s": 14.21, "r_error": "", "src_len": 5843, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6674, "ts": 1791535210.1274364, "chk": "no_custom_kernel", "code_len": 5878}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5557, "tout": 3064, "ts": 1791535248.0641172, "chk": "ok", "code_len": 5843}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8480, "tout": 2761, "ts": 1791535285.2097962, "chk": null, "code_len": 5843}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2183, "started": 1791535150.9177773}
{"ep": "h2-00557", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0287, "speedup": 1.0287, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 110.2, "calls": 1, "tokens_in": 5832, "tokens_out": 524, "ctx_chars": 9656, "t_end": 1791535323.2964118, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0287, "lo95": 1.0273, "mean": 1.0284, "ev_only": 1.0288, "ref_ms": 24.85754, "cand_ms": 24.1645, "n_trials": 54, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0286, "r_wall_s": 104.75, "compile_s": 2.18, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 524, "ts": 1791535217.8586564, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791535213.140492}
{"ep": "h2-00559", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0115, "speedup": 1.0115, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 93.6, "calls": 3, "tokens_in": 19805, "tokens_out": 2073, "ctx_chars": 9274, "t_end": 1791535325.2331603, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0115, "lo95": 1.0104, "mean": 1.012, "ev_only": 1.0139, "ref_ms": 8.3324, "cand_ms": 8.23787, "n_trials": 149, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.015, "r_wall_s": 47.72, "compile_s": 1.94, "r_error": "", "src_len": 1962, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1087, "ts": 1791535242.8511558, "chk": "ok", "code_len": 2005}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6989, "tout": 13, "ts": 1791535269.3435237, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7026, "tout": 973, "ts": 1791535277.436919, "chk": null, "code_len": 1962}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535232.4617915}
{"ep": "h2-00481", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.6643, "speedup": 2.6643, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 369.4, "calls": 2, "tokens_in": 13006, "tokens_out": 4188, "ctx_chars": 8529, "t_end": 1791535326.4333868, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 2.6643, "lo95": 2.6618, "mean": 2.6626, "ev_only": 2.6742, "ref_ms": 8.52686, "cand_ms": 3.20037, "n_trials": 177, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1406, "r_wall_s": 46.14, "compile_s": 1.97, "r_error": "", "src_len": 2929, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2579, "ts": 1791534986.1328943, "chk": "ok", "code_len": 2929}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7509, "tout": 1609, "ts": 1791535280.259479, "chk": null, "code_len": 2929}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791534957.0097418}
{"ep": "h2-00533", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 3.0042, "speedup": 3.0042, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 218.2, "calls": 2, "tokens_in": 15405, "tokens_out": 5539, "ctx_chars": 11065, "t_end": 1791535329.2568898, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 3.0042, "lo95": 2.9851, "mean": 3.0605, "ev_only": 3.1953, "ref_ms": 1.83567, "cand_ms": 0.61103, "n_trials": 218, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 71.46, "compile_s": 1.93, "r_error": "", "src_len": 3116, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 3993, "ts": 1791535148.0716476, "chk": "ok", "code_len": 3116}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8776, "tout": 1546, "ts": 1791535257.749675, "chk": null, "code_len": 3116}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791535111.076914}
{"ep": "h2-00508", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0064, "speedup": 1.0064, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 313.8, "calls": 2, "tokens_in": 4043, "tokens_out": 2190, "ctx_chars": 0, "t_end": 1791535329.9649053, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0064, "lo95": 1.0055, "mean": 1.0081, "ev_only": 1.0071, "ref_ms": 8.00688, "cand_ms": 7.95574, "n_trials": 59, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 45.24, "compile_s": 2.45, "r_error": "", "src_len": 997, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1596, "ts": 1791535032.2891674, "chk": "ok", "code_len": 997}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2540, "tout": 594, "ts": 1791535284.6558712, "chk": null, "code_len": 997}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791535016.1355307}
{"ep": "h2-00561", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.9982, "speedup": 0.9982, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.8, "calls": 2, "tokens_in": 4160, "tokens_out": 1946, "ctx_chars": 0, "t_end": 1791535331.106959, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9982, "lo95": 0.9958, "mean": 1.0008, "ev_only": 1.0015, "ref_ms": 8.0208, "cand_ms": 8.03487, "n_trials": 144, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 61.02, "compile_s": 2.0, "r_error": "", "src_len": 906, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1365, "ts": 1791535246.259601, "chk": "ok", "code_len": 906}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2559, "tout": 581, "ts": 1791535269.9864223, "chk": null, "code_len": 906}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791535233.3075507}
{"ep": "h2-00574", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 69.1, "calls": 1, "tokens_in": 2743, "tokens_out": 2470, "ctx_chars": 0, "t_end": 1791535332.351947, "error": "RuntimeError: Error building extension 'resnet_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=resnet_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 46.2, "compile_s": 46.19, "r_error": "RuntimeError: Error building extension 'resnet_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=resnet_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "src_len": 5148, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2743, "tout": 2470, "ts": 1791535286.1067379, "chk": null, "code_len": 5148}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2743, "started": 1791535263.2597141}
{"ep": "h2-00433", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.7777, "speedup": 2.7777, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 526.6, "calls": 3, "tokens_in": 13714, "tokens_out": 9957, "ctx_chars": 0, "t_end": 1791535334.5845096, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.7777, "lo95": 2.7763, "mean": 2.7737, "ev_only": 2.7867, "ref_ms": 7.8185, "cand_ms": 2.81476, "n_trials": 190, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9987, "r_wall_s": 346.49, "compile_s": 15.12, "r_error": "", "src_len": 4832, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 4916, "ts": 1791534848.7756329, "chk": "no_custom_kernel", "code_len": 4852}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4705, "tout": 2650, "ts": 1791534902.4446328, "chk": "ok", "code_len": 4832}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7284, "tout": 2391, "ts": 1791534987.1402085, "chk": null, "code_len": 4832}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1725, "started": 1791534807.978412}
{"ep": "h2-00564", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0594, "speedup": 1.0594, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 91.9, "calls": 3, "tokens_in": 25400, "tokens_out": 4290, "ctx_chars": 10167, "t_end": 1791535336.4823318, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0594, "lo95": 1.0589, "mean": 1.0598, "ev_only": 1.0618, "ref_ms": 8.43802, "cand_ms": 7.96469, "n_trials": 139, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 20.78, "compile_s": 1.94, "r_error": "", "src_len": 1663, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3386, "ts": 1791535284.3643222, "chk": "ok", "code_len": 1663}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9602, "tout": 68, "ts": 1791535308.0519693, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9694, "tout": 836, "ts": 1791535315.5361462, "chk": null, "code_len": 1663}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791535244.5442076}
{"ep": "h2-00511", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0091, "speedup": 1.0091, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 318.7, "calls": 2, "tokens_in": 11624, "tokens_out": 1328, "ctx_chars": 9028, "t_end": 1791535337.3311472, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0091, "lo95": 1.0062, "mean": 1.0059, "ev_only": 1.0061, "ref_ms": 8.07902, "cand_ms": 8.00589, "n_trials": 152, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0076, "r_wall_s": 69.4, "compile_s": 1.87, "r_error": "", "src_len": 977, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 770, "ts": 1791535025.3167095, "chk": "ok", "code_len": 977}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6253, "tout": 558, "ts": 1791535267.835794, "chk": null, "code_len": 977}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791535018.585743}
{"ep": "h2-00585", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8496, "speedup": 0.8496, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 35.8, "calls": 1, "tokens_in": 7030, "tokens_out": 1203, "ctx_chars": 11175, "t_end": 1791535341.8677433, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.8496, "lo95": 0.846, "mean": 0.8476, "ev_only": 0.8471, "ref_ms": 14.40386, "cand_ms": 16.95411, "n_trials": 75, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 24.93, "compile_s": 1.98, "r_error": "", "src_len": 2527, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1203, "ts": 1791535316.7670085, "chk": null, "code_len": 2527}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791535306.0815783}
{"ep": "h2-00540", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.0125, "speedup": 0.0125, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 200.5, "calls": 2, "tokens_in": 16690, "tokens_out": 5757, "ctx_chars": 11539, "t_end": 1791535353.775985, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.0125, "lo95": 0.0125, "mean": 0.0125, "ev_only": 0.0125, "ref_ms": 7.79026, "cand_ms": 622.76022, "n_trials": 12, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 101.33, "compile_s": 1.9, "r_error": "", "src_len": 3857, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 3768, "ts": 1791535187.3068068, "chk": "ok", "code_len": 3857}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9575, "tout": 1989, "ts": 1791535251.8265314, "chk": null, "code_len": 3857}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7115, "started": 1791535153.3156204}
{"ep": "h2-00414", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.7089, "speedup": 2.7089, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 651.4, "calls": 2, "tokens_in": 15505, "tokens_out": 38578, "ctx_chars": 0, "t_end": 1791535354.8388596, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.7089, "lo95": 2.6994, "mean": 2.7229, "ev_only": 2.7056, "ref_ms": 6.893, "cand_ms": 2.54456, "n_trials": 52, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.003, "r_wall_s": 321.55, "compile_s": 19.4, "r_error": "", "src_len": 18846, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 29605, "ts": 1791534954.9504054, "chk": "ok", "code_len": 18846}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 12561, "tout": 8973, "ts": 1791535032.4642582, "chk": null, "code_len": 18846}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791534703.4301896}
{"ep": "h2-00589", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7829, "speedup": 0.7829, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 42.8, "calls": 3, "tokens_in": 18948, "tokens_out": 1612, "ctx_chars": 8224, "t_end": 1791535357.011875, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.7829, "lo95": 0.7808, "mean": 0.8445, "ev_only": 0.7826, "ref_ms": 59.68603, "cand_ms": 76.23689, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 14.16, "compile_s": 2.04, "r_error": "", "src_len": 1070, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 1028, "ts": 1791535326.65801, "chk": "ok", "code_len": 1070}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6665, "tout": 66, "ts": 1791535338.0260127, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6755, "tout": 518, "ts": 1791535342.796834, "chk": null, "code_len": 1070}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791535314.178388}
{"ep": "h2-00556", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0556, "speedup": 1.0556, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 147.2, "calls": 2, "tokens_in": 8387, "tokens_out": 6481, "ctx_chars": 0, "t_end": 1791535358.154074, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0556, "lo95": 1.0498, "mean": 1.056, "ev_only": 1.0578, "ref_ms": 3.30709, "cand_ms": 3.13304, "n_trials": 197, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1078, "r_wall_s": 70.49, "compile_s": 1.88, "r_error": "", "src_len": 5512, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 3946, "ts": 1791535242.990904, "chk": "ok", "code_len": 5512}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5642, "tout": 2535, "ts": 1791535287.6219864, "chk": null, "code_len": 5512}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791535210.9771163}
{"ep": "h2-00538", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9509, "speedup": 0.9509, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 225.5, "calls": 2, "tokens_in": 14039, "tokens_out": 11914, "ctx_chars": 6513, "t_end": 1791535361.906559, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9509, "lo95": 0.95, "mean": 0.9521, "ev_only": 0.9512, "ref_ms": 15.71305, "cand_ms": 16.52423, "n_trials": 64, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.5706, "r_wall_s": 72.72, "compile_s": 1.91, "r_error": "", "src_len": 7636, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 8171, "ts": 1791535223.5126102, "chk": "ok", "code_len": 7636}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9175, "tout": 3743, "ts": 1791535288.965562, "chk": null, "code_len": 7636}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535136.4060185}
{"ep": "h2-00573", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0183, "speedup": 1.0183, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.1, "calls": 1, "tokens_in": 1750, "tokens_out": 2385, "ctx_chars": 0, "t_end": 1791535362.2168067, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0183, "lo95": 1.0178, "mean": 1.0198, "ev_only": 1.0191, "ref_ms": 8.13571, "cand_ms": 7.98979, "n_trials": 137, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0096, "r_wall_s": 59.03, "compile_s": 39.19, "r_error": "", "src_len": 1974, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1750, "tout": 2385, "ts": 1791535303.0957763, "chk": null, "code_len": 1974}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1750, "started": 1791535262.0725455}
{"ep": "h2-00578", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8143, "speedup": 0.8143, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 91.2, "calls": 1, "tokens_in": 1984, "tokens_out": 817, "ctx_chars": 0, "t_end": 1791535362.76161, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.8143, "lo95": 0.8113, "mean": 0.8132, "ev_only": 0.8148, "ref_ms": 60.43204, "cand_ms": 74.21183, "n_trials": 17, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 81.67, "compile_s": 1.93, "r_error": "", "src_len": 926, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1984, "tout": 817, "ts": 1791535281.0564978, "chk": null, "code_len": 926}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1984, "started": 1791535271.600929}
{"ep": "h2-00560", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0142, "speedup": 1.0142, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 132.2, "calls": 2, "tokens_in": 11422, "tokens_out": 4410, "ctx_chars": 5026, "t_end": 1791535363.9340377, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0142, "lo95": 1.0132, "mean": 1.0148, "ev_only": 1.0164, "ref_ms": 8.14394, "cand_ms": 8.0297, "n_trials": 147, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0087, "r_wall_s": 85.45, "compile_s": 2.17, "r_error": "", "src_len": 2108, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 3324, "ts": 1791535259.4757445, "chk": "ok", "code_len": 2108}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7429, "tout": 1086, "ts": 1791535278.3166032, "chk": null, "code_len": 2108}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791535231.7645202}
{"ep": "h2-00499", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 3.2788, "speedup": 3.2788, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 368.4, "calls": 2, "tokens_in": 15469, "tokens_out": 5616, "ctx_chars": 11065, "t_end": 1791535366.9304185, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 3.2788, "lo95": 3.2642, "mean": 3.2619, "ev_only": 3.3672, "ref_ms": 1.81101, "cand_ms": 0.55233, "n_trials": 246, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 19.17, "compile_s": 2.07, "r_error": "", "src_len": 3095, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 3914, "ts": 1791535035.1265213, "chk": "ok", "code_len": 3095}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8840, "tout": 1702, "ts": 1791535347.7353587, "chk": null, "code_len": 3095}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791534998.488384}
{"ep": "h2-00562", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0705, "speedup": 1.0705, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 129.2, "calls": 2, "tokens_in": 4643, "tokens_out": 2300, "ctx_chars": 0, "t_end": 1791535367.5694556, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0705, "lo95": 1.0384, "mean": 1.0464, "ev_only": 0.9857, "ref_ms": 0.16018, "cand_ms": 0.14963, "n_trials": 222, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 92.93, "compile_s": 11.69, "r_error": "", "src_len": 1916, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 1236, "ts": 1791535248.138426, "chk": "ok", "code_len": 1916}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2994, "tout": 1064, "ts": 1791535274.4993231, "chk": null, "code_len": 1916}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791535238.3213139}
{"ep": "h2-00572", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.2009, "speedup": 0.2009, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 113.7, "calls": 2, "tokens_in": 15362, "tokens_out": 6430, "ctx_chars": 11065, "t_end": 1791535372.7209737, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.2009, "lo95": 0.2006, "mean": 0.2012, "ev_only": 0.2, "ref_ms": 1.81593, "cand_ms": 9.04098, "n_trials": 184, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 29.73, "compile_s": 1.94, "r_error": "", "src_len": 3092, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 4894, "ts": 1791535313.303754, "chk": "ok", "code_len": 3092}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8733, "tout": 1536, "ts": 1791535342.8682768, "chk": null, "code_len": 3092}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791535259.0431504}
{"ep": "h2-00488", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8452, "speedup": 0.8452, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 394.2, "calls": 2, "tokens_in": 15814, "tokens_out": 3203, "ctx_chars": 11175, "t_end": 1791535372.790493, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8452, "lo95": 0.8442, "mean": 0.845, "ev_only": 0.8453, "ref_ms": 14.31103, "cand_ms": 16.93119, "n_trials": 60, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 34.18, "compile_s": 1.92, "r_error": "", "src_len": 2406, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1879, "ts": 1791534994.0776463, "chk": "ok", "code_len": 2406}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8782, "tout": 1324, "ts": 1791535338.5230427, "chk": null, "code_len": 2406}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791534978.5422719}
{"ep": "h2-00595", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9995, "speedup": 0.9995, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 43.9, "calls": 1, "tokens_in": 2181, "tokens_out": 983, "ctx_chars": 0, "t_end": 1791535373.1429718, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9995, "lo95": 0.9989, "mean": 1.036, "ev_only": 0.9986, "ref_ms": 14.32992, "cand_ms": 14.33666, "n_trials": 100, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0021, "r_wall_s": 32.03, "compile_s": 3.14, "r_error": "", "src_len": 1115, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2181, "tout": 983, "ts": 1791535340.992566, "chk": null, "code_len": 1115}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2181, "started": 1791535329.2573392}
{"ep": "h2-00555", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.646, "speedup": 2.646, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 169.8, "calls": 2, "tokens_in": 13048, "tokens_out": 4195, "ctx_chars": 8529, "t_end": 1791535375.3658655, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.646, "lo95": 2.6451, "mean": 2.6433, "ev_only": 2.6555, "ref_ms": 8.49246, "cand_ms": 3.2096, "n_trials": 153, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3453, "r_wall_s": 98.05, "compile_s": 2.27, "r_error": "", "src_len": 2839, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2657, "ts": 1791535229.211834, "chk": "ok", "code_len": 2839}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7551, "tout": 1538, "ts": 1791535277.2575006, "chk": null, "code_len": 2839}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791535205.5589588}
{"ep": "h2-00584", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0085, "speedup": 1.0085, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.1, "calls": 2, "tokens_in": 12742, "tokens_out": 2077, "ctx_chars": 9274, "t_end": 1791535375.8372095, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0085, "lo95": 1.0069, "mean": 1.0092, "ev_only": 1.006, "ref_ms": 8.39042, "cand_ms": 8.3199, "n_trials": 138, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0429, "r_wall_s": 38.71, "compile_s": 2.28, "r_error": "", "src_len": 1969, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1050, "ts": 1791535308.412647, "chk": "ok", "code_len": 1969}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6952, "tout": 1027, "ts": 1791535337.0266635, "chk": null, "code_len": 1969}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5790, "started": 1791535298.755351}
{"ep": "h2-00507", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0049, "speedup": 1.0049, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 368.7, "calls": 3, "tokens_in": 19849, "tokens_out": 2071, "ctx_chars": 9274, "t_end": 1791535379.4355102, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0049, "lo95": 1.0043, "mean": 1.0062, "ev_only": 1.0036, "ref_ms": 8.30084, "cand_ms": 8.26077, "n_trials": 172, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0421, "r_wall_s": 45.19, "compile_s": 2.47, "r_error": "", "src_len": 1850, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1109, "ts": 1791535021.2680752, "chk": "ok", "code_len": 1899}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7011, "tout": 13, "ts": 1791535324.595313, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7048, "tout": 949, "ts": 1791535334.0679197, "chk": null, "code_len": 1850}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535010.7676096}
{"ep": "h2-00558", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.199, "speedup": 1.199, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 157.2, "calls": 3, "tokens_in": 9728, "tokens_out": 3855, "ctx_chars": 0, "t_end": 1791535382.4719286, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.199, "lo95": 1.1962, "mean": 1.207, "ev_only": 1.1975, "ref_ms": 1.81357, "cand_ms": 1.51254, "n_trials": 191, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1452, "r_wall_s": 92.61, "compile_s": 2.43, "r_error": "", "src_len": 3790, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1757, "tout": 2098, "ts": 1791535245.6347244, "chk": "ok", "code_len": 3824}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3967, "tout": 13, "ts": 1791535275.6783612, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 4004, "tout": 1744, "ts": 1791535289.8368776, "chk": null, "code_len": 3790}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1757, "started": 1791535225.247999}
{"ep": "h2-00594", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.634, "speedup": 0.634, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 58.8, "calls": 1, "tokens_in": 1647, "tokens_out": 449, "ctx_chars": 0, "t_end": 1791535385.2475488, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.634, "lo95": 0.6126, "mean": 0.6937, "ev_only": 0.6162, "ref_ms": 0.16697, "cand_ms": 0.26337, "n_trials": 190, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 53.36, "compile_s": 2.01, "r_error": "", "src_len": 802, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1647, "tout": 449, "ts": 1791535331.6766787, "chk": null, "code_len": 802}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1647, "started": 1791535326.4336216}
{"ep": "h2-00475", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0657, "speedup": 1.0657, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 442.6, "calls": 2, "tokens_in": 10501, "tokens_out": 2302, "ctx_chars": 6513, "t_end": 1791535385.2922564, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0657, "lo95": 1.0649, "mean": 1.0644, "ev_only": 1.0649, "ref_ms": 15.48551, "cand_ms": 14.53096, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1172, "r_wall_s": 365.47, "compile_s": 4.07, "r_error": "", "src_len": 2423, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1179, "ts": 1791534953.3970916, "chk": "ok", "code_len": 2423}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5896, "tout": 1123, "ts": 1791535018.5829268, "chk": null, "code_len": 2423}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791534942.7380748}
{"ep": "h2-00576", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0631, "speedup": 1.0631, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 119.2, "calls": 2, "tokens_in": 9154, "tokens_out": 8327, "ctx_chars": 0, "t_end": 1791535387.2687507, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0631, "lo95": 1.0611, "mean": 1.0655, "ev_only": 1.0661, "ref_ms": 3.32709, "cand_ms": 3.12948, "n_trials": 187, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0929, "r_wall_s": 41.54, "compile_s": 2.96, "r_error": "", "src_len": 6918, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 5041, "ts": 1791535309.5142682, "chk": "ok", "code_len": 6918}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6409, "tout": 3286, "ts": 1791535345.6859968, "chk": null, "code_len": 6918}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791535268.0455909}
{"ep": "h2-00597", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0003, "speedup": 1.0003, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 61.3, "calls": 2, "tokens_in": 13440, "tokens_out": 1112, "ctx_chars": 10847, "t_end": 1791535392.383671, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0003, "lo95": 0.9995, "mean": 1.0003, "ev_only": 1.0008, "ref_ms": 8.02038, "cand_ms": 8.01783, "n_trials": 161, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 38.96, "compile_s": 2.12, "r_error": "", "src_len": 823, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 593, "ts": 1791535336.3512065, "chk": "ok", "code_len": 823}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7071, "tout": 519, "ts": 1791535353.3211308, "chk": null, "code_len": 823}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791535331.1072025}
{"ep": "h2-00504", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8445, "speedup": 0.8445, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 392.8, "calls": 1, "tokens_in": 7030, "tokens_out": 1203, "ctx_chars": 11175, "t_end": 1791535398.6022785, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8445, "lo95": 0.8438, "mean": 0.8448, "ev_only": 0.8451, "ref_ms": 14.30267, "cand_ms": 16.93555, "n_trials": 89, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 381.63, "compile_s": 6.29, "r_error": "", "src_len": 2534, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1203, "ts": 1791535016.6063974, "chk": null, "code_len": 2534}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791535005.8323963}
{"ep": "h2-00496", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0001, "speedup": 1.0001, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 404.1, "calls": 1, "tokens_in": 5369, "tokens_out": 564, "ctx_chars": 9028, "t_end": 1791535399.7527075, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0001, "lo95": 0.9931, "mean": 1.0007, "ev_only": 1.0022, "ref_ms": 8.10495, "cand_ms": 8.10392, "n_trials": 79, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 398.46, "compile_s": 2.94, "r_error": "", "src_len": 1046, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5369, "tout": 564, "ts": 1791535000.9736588, "chk": null, "code_len": 1046}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5369, "started": 1791534995.608191}
{"ep": "h2-00468", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4119, "speedup": 1.4119, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 478.1, "calls": 2, "tokens_in": 5201, "tokens_out": 5042, "ctx_chars": 0, "t_end": 1791535401.724528, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.4119, "lo95": 1.4115, "mean": 1.4134, "ev_only": 1.4104, "ref_ms": 15.51796, "cand_ms": 10.99099, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9721, "r_wall_s": 410.12, "compile_s": 2.57, "r_error": "", "src_len": 2476, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 3455, "ts": 1791534953.9194307, "chk": "ok", "code_len": 2476}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3634, "tout": 1587, "ts": 1791534991.0510654, "chk": null, "code_len": 2476}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1567, "started": 1791534923.6551547}
{"ep": "h2-00583", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9913, "speedup": 0.9913, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 110.6, "calls": 2, "tokens_in": 15575, "tokens_out": 2157, "ctx_chars": 11539, "t_end": 1791535402.4063604, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9913, "lo95": 0.9907, "mean": 0.9905, "ev_only": 0.9899, "ref_ms": 7.75787, "cand_ms": 7.82598, "n_trials": 149, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0102, "r_wall_s": 69.03, "compile_s": 2.32, "r_error": "", "src_len": 1821, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1240, "ts": 1791535305.4108386, "chk": "ok", "code_len": 1821}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8462, "tout": 917, "ts": 1791535332.6519294, "chk": null, "code_len": 1821}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7113, "started": 1791535292.3617737}
{"ep": "h2-00493", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0141, "speedup": 1.0141, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 421.3, "calls": 3, "tokens_in": 24633, "tokens_out": 2103, "ctx_chars": 12117, "t_end": 1791535405.5281608, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0141, "lo95": 1.0128, "mean": 1.0162, "ev_only": 1.0142, "ref_ms": 42.71521, "cand_ms": 42.12234, "n_trials": 16, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0, "r_wall_s": 117.73, "compile_s": 2.22, "r_error": "", "src_len": 2157, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1074, "ts": 1791534994.2998827, "chk": "ok", "code_len": 2157}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8583, "tout": 46, "ts": 1791535278.5490942, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8653, "tout": 983, "ts": 1791535287.1156087, "chk": null, "code_len": 2157}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791534984.1967432}
{"ep": "h2-00571", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8476, "speedup": 0.8476, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 153.1, "calls": 1, "tokens_in": 7030, "tokens_out": 1207, "ctx_chars": 11175, "t_end": 1791535411.5119667, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.8476, "lo95": 0.8454, "mean": 0.8749, "ev_only": 0.8485, "ref_ms": 14.32629, "cand_ms": 16.90222, "n_trials": 56, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 127.15, "compile_s": 2.2, "r_error": "", "src_len": 2551, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1207, "ts": 1791535284.27204, "chk": null, "code_len": 2551}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791535258.364679}
{"ep": "h2-00566", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.02, "speedup": 1.02, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 165.9, "calls": 2, "tokens_in": 10190, "tokens_out": 3172, "ctx_chars": 5026, "t_end": 1791535415.7056272, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.02, "lo95": 1.0193, "mean": 1.0912, "ev_only": 1.0199, "ref_ms": 8.14772, "cand_ms": 7.98809, "n_trials": 77, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0057, "r_wall_s": 69.65, "compile_s": 2.74, "r_error": "", "src_len": 2097, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 2092, "ts": 1791535285.7492547, "chk": "ok", "code_len": 2097}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6197, "tout": 1080, "ts": 1791535345.712217, "chk": null, "code_len": 2097}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791535249.8276074}
{"ep": "h2-00474", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.4153, "speedup": 3.4153, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 473.6, "calls": 3, "tokens_in": 15908, "tokens_out": 13485, "ctx_chars": 0, "t_end": 1791535415.8530035, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 3.4153, "lo95": 3.3962, "mean": 3.5638, "ev_only": 3.4226, "ref_ms": 61.32909, "cand_ms": 17.95729, "n_trials": 12, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8595, "r_wall_s": 86.01, "compile_s": 13.84, "r_error": "", "src_len": 5784, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 7646, "ts": 1791535011.5596666, "chk": "no_custom_kernel", "code_len": 5806}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5475, "tout": 3064, "ts": 1791535293.759545, "chk": "ok", "code_len": 5784}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8447, "tout": 2775, "ts": 1791535329.7262073, "chk": null, "code_len": 5784}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791534942.2888951}
{"ep": "h2-00553", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0263, "speedup": 1.0263, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 216.9, "calls": 2, "tokens_in": 4392, "tokens_out": 2205, "ctx_chars": 0, "t_end": 1791535416.9790416, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0263, "lo95": 1.0242, "mean": 1.0256, "ev_only": 1.0268, "ref_ms": 24.93602, "cand_ms": 24.29595, "n_trials": 35, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0291, "r_wall_s": 129.91, "compile_s": 2.34, "r_error": "", "src_len": 1065, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1521, "ts": 1791535220.5340545, "chk": "ok", "code_len": 1065}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2724, "tout": 684, "ts": 1791535286.3814676, "chk": null, "code_len": 1065}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791535200.0915654}
{"ep": "h2-00602", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 78.4, "calls": 1, "tokens_in": 2159, "tokens_out": 2077, "ctx_chars": 0, "t_end": 1791535420.251513, "error": "RuntimeError: Error building extension 'mbconv_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mbconv_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 59.81, "compile_s": 59.8, "r_error": "RuntimeError: Error building extension 'mbconv_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mbconv_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "src_len": 4269, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2159, "tout": 2077, "ts": 1791535360.4199586, "chk": null, "code_len": 4269}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2159, "started": 1791535341.8680012}
{"ep": "h2-00603", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.5542, "speedup": 0.5542, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 85.5, "calls": 1, "tokens_in": 1565, "tokens_out": 1862, "ctx_chars": 0, "t_end": 1791535439.2675173, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.5542, "lo95": 0.5538, "mean": 0.5541, "ev_only": 0.5535, "ref_ms": 15.49751, "cand_ms": 27.96336, "n_trials": 40, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 68.49, "compile_s": 2.12, "r_error": "", "src_len": 3082, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1565, "tout": 1862, "ts": 1791535370.4031456, "chk": null, "code_len": 3082}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1565, "started": 1791535354.3111835}
{"ep": "h2-00575", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.3791, "speedup": 1.3791, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 179.9, "calls": 2, "tokens_in": 5277, "tokens_out": 4345, "ctx_chars": 0, "t_end": 1791535444.8749382, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3791, "lo95": 1.3718, "mean": 1.3919, "ev_only": 1.3927, "ref_ms": 7.98144, "cand_ms": 5.78759, "n_trials": 87, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 84.8, "compile_s": 1.94, "r_error": "", "src_len": 2832, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 2755, "ts": 1791535289.455555, "chk": "ok", "code_len": 2832}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3653, "tout": 1590, "ts": 1791535359.9813507, "chk": null, "code_len": 2832}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791535265.0128021}
{"ep": "h2-00472", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4618, "speedup": 1.4618, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 508.8, "calls": 2, "tokens_in": 11313, "tokens_out": 5636, "ctx_chars": 6513, "t_end": 1791535447.8736002, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.4618, "lo95": 1.4612, "mean": 1.4586, "ev_only": 1.4631, "ref_ms": 15.5, "cand_ms": 10.60347, "n_trials": 92, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1908, "r_wall_s": 441.85, "compile_s": 3.61, "r_error": "", "src_len": 3350, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 4065, "ts": 1791534976.2668154, "chk": "ok", "code_len": 3350}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6706, "tout": 1571, "ts": 1791535005.0758252, "chk": null, "code_len": 3350}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791534939.082719}
{"ep": "h2-00623", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7153, "speedup": 0.7153, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 60.3, "calls": 2, "tokens_in": 16667, "tokens_out": 4635, "ctx_chars": 12352, "t_end": 1791535452.7240715, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.7153, "lo95": 0.715, "mean": 0.7158, "ev_only": 0.7154, "ref_ms": 64.58892, "cand_ms": 90.29004, "n_trials": 15, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.002, "r_wall_s": 12.91, "compile_s": 1.97, "r_error": "", "src_len": 2574, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 3316, "ts": 1791535422.2978258, "chk": "ok", "code_len": 2574}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9337, "tout": 1319, "ts": 1791535439.789556, "chk": null, "code_len": 2574}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791535392.383945}
{"ep": "h2-00626", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.018, "speedup": 1.018, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 53.9, "calls": 3, "tokens_in": 23780, "tokens_out": 2501, "ctx_chars": 11823, "t_end": 1791535455.6238441, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.018, "lo95": 1.0169, "mean": 1.0185, "ev_only": 1.0182, "ref_ms": 27.52212, "cand_ms": 27.03675, "n_trials": 43, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0145, "r_wall_s": 20.09, "compile_s": 1.88, "r_error": "", "src_len": 2120, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1358, "ts": 1791535416.5869408, "chk": "ok", "code_len": 2362}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8391, "tout": 53, "ts": 1791535426.209889, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8468, "tout": 1090, "ts": 1791535435.508791, "chk": null, "code_len": 2120}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6921, "started": 1791535401.7249062}
{"ep": "h2-00629", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0177, "speedup": 1.0177, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 56.2, "calls": 2, "tokens_in": 9146, "tokens_out": 1795, "ctx_chars": 5026, "t_end": 1791535467.6864295, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0177, "lo95": 1.0174, "mean": 1.0256, "ev_only": 1.0197, "ref_ms": 8.12426, "cand_ms": 7.98332, "n_trials": 157, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0089, "r_wall_s": 16.87, "compile_s": 1.88, "r_error": "", "src_len": 1442, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1048, "ts": 1791535436.1669974, "chk": "ok", "code_len": 1442}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5153, "tout": 747, "ts": 1791535450.7295637, "chk": null, "code_len": 1442}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791535411.5122316}
{"ep": "h2-00565", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7822, "speedup": 0.7822, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 226.2, "calls": 2, "tokens_in": 16880, "tokens_out": 7577, "ctx_chars": 11539, "t_end": 1791535472.7501998, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7822, "lo95": 0.7796, "mean": 0.7848, "ev_only": 0.7805, "ref_ms": 7.80799, "cand_ms": 9.98202, "n_trials": 101, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 102.75, "compile_s": 2.04, "r_error": "", "src_len": 4395, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 5289, "ts": 1791535326.1186638, "chk": "ok", "code_len": 4395}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9765, "tout": 2288, "ts": 1791535369.268218, "chk": null, "code_len": 4395}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7115, "started": 1791535246.5109465}
{"ep": "h2-00592", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0167, "speedup": 1.0167, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 150.1, "calls": 3, "tokens_in": 24697, "tokens_out": 2223, "ctx_chars": 12352, "t_end": 1791535473.4079666, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0167, "lo95": 1.0149, "mean": 1.0166, "ev_only": 1.0168, "ref_ms": 64.50162, "cand_ms": 63.44473, "n_trials": 11, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 41.81, "compile_s": 2.09, "r_error": "", "src_len": 1915, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1226, "ts": 1791535335.3022826, "chk": "ok", "code_len": 1915}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8666, "tout": 13, "ts": 1791535423.06146, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8703, "tout": 984, "ts": 1791535431.5575438, "chk": null, "code_len": 1915}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791535323.2966664}
{"ep": "h2-00500", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0062, "speedup": 1.0062, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 474.8, "calls": 3, "tokens_in": 19941, "tokens_out": 2710, "ctx_chars": 9266, "t_end": 1791535475.5032516, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0062, "lo95": 1.0053, "mean": 1.006, "ev_only": 1.0052, "ref_ms": 30.45647, "cand_ms": 30.26925, "n_trials": 21, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.041, "r_wall_s": 135.22, "compile_s": 2.47, "r_error": "", "src_len": 2611, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1401, "ts": 1791535014.1409316, "chk": "ok", "code_len": 2611}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7127, "tout": 49, "ts": 1791535328.1489081, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7200, "tout": 1260, "ts": 1791535339.72898, "chk": null, "code_len": 2611}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791535000.6617858}
{"ep": "h2-00577", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.4071, "speedup": 1.4071, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 208.0, "calls": 2, "tokens_in": 30845, "tokens_out": 18025, "ctx_chars": 2425, "t_end": 1791535479.5204756, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.4071, "lo95": 1.4034, "mean": 1.4084, "ev_only": 1.4064, "ref_ms": 19.75923, "cand_ms": 14.04272, "n_trials": 79, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3808, "r_wall_s": 43.07, "compile_s": 1.85, "r_error": "", "src_len": 16383, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 10741, "ts": 1791535353.4515643, "chk": "ok", "code_len": 16383}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19295, "tout": 7284, "ts": 1791535436.4210966, "chk": null, "code_len": 16383}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791535271.5265415}
{"ep": "h2-00618", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0065, "speedup": 1.0065, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.3, "calls": 3, "tokens_in": 19871, "tokens_out": 2148, "ctx_chars": 9274, "t_end": 1791535479.728046, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0065, "lo95": 1.0061, "mean": 1.0068, "ev_only": 1.0069, "ref_ms": 8.28959, "cand_ms": 8.23586, "n_trials": 143, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0419, "r_wall_s": 19.29, "compile_s": 1.89, "r_error": "", "src_len": 1981, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1099, "ts": 1791535389.6399314, "chk": "ok", "code_len": 2046}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7001, "tout": 55, "ts": 1791535452.0733035, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7080, "tout": 994, "ts": 1791535460.344119, "chk": null, "code_len": 1981}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535379.43578}
{"ep": "h2-00598", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9766, "speedup": 0.9766, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 149.1, "calls": 1, "tokens_in": 7032, "tokens_out": 1135, "ctx_chars": 11539, "t_end": 1791535481.44699, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9766, "lo95": 0.9764, "mean": 0.978, "ev_only": 0.9761, "ref_ms": 31.36993, "cand_ms": 32.12246, "n_trials": 39, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0238, "r_wall_s": 136.62, "compile_s": 2.28, "r_error": "", "src_len": 1798, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1135, "ts": 1791535343.9572437, "chk": null, "code_len": 1798}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791535332.3521829}
{"ep": "h2-00599", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.5507, "speedup": 0.5507, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 148.2, "calls": 3, "tokens_in": 20425, "tokens_out": 3558, "ctx_chars": 9033, "t_end": 1791535482.7542841, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.5507, "lo95": 0.5505, "mean": 0.5507, "ev_only": 0.5509, "ref_ms": 15.47213, "cand_ms": 28.0959, "n_trials": 48, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 50.7, "compile_s": 2.1, "r_error": "", "src_len": 2861, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1672, "ts": 1791535350.4211993, "chk": "ok", "code_len": 2861}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7269, "tout": 375, "ts": 1791535418.8590515, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7668, "tout": 1511, "ts": 1791535431.589592, "chk": null, "code_len": 2861}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5488, "started": 1791535334.584743}
{"ep": "h2-00550", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0699, "speedup": 1.0699, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 298.8, "calls": 2, "tokens_in": 14429, "tokens_out": 19471, "ctx_chars": 6513, "t_end": 1791535487.8410602, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0699, "lo95": 1.0695, "mean": 1.0682, "ev_only": 1.07, "ref_ms": 15.69798, "cand_ms": 14.67229, "n_trials": 71, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4685, "r_wall_s": 105.73, "compile_s": 5.5, "r_error": "", "src_len": 8060, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 15444, "ts": 1791535333.0645642, "chk": "ok", "code_len": 8060}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9565, "tout": 4027, "ts": 1791535381.9487848, "chk": null, "code_len": 8060}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535189.0909932}
{"ep": "h2-00633", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8398, "speedup": 0.8398, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 68.6, "calls": 2, "tokens_in": 15382, "tokens_out": 2374, "ctx_chars": 11175, "t_end": 1791535488.8293548, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8398, "lo95": 0.838, "mean": 0.8388, "ev_only": 0.8392, "ref_ms": 14.33177, "cand_ms": 17.06644, "n_trials": 67, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 37.62, "compile_s": 2.08, "r_error": "", "src_len": 2493, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7030, "tout": 1210, "ts": 1791535432.580661, "chk": "ok", "code_len": 2493}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8352, "tout": 1164, "ts": 1791535451.1179488, "chk": null, "code_len": 2493}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7030, "started": 1791535420.2517452}
{"ep": "h2-00635", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0041, "speedup": 1.0041, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 45.4, "calls": 3, "tokens_in": 17460, "tokens_out": 1107, "ctx_chars": 9028, "t_end": 1791535490.3142052, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0041, "lo95": 1.0035, "mean": 1.004, "ev_only": 1.0028, "ref_ms": 7.97776, "cand_ms": 7.9455, "n_trials": 87, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 18.99, "compile_s": 2.08, "r_error": "", "src_len": 1024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 546, "ts": 1791535451.2438915, "chk": "ok", "code_len": 1024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6027, "tout": 13, "ts": 1791535466.0324597, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6064, "tout": 548, "ts": 1791535471.2276769, "chk": null, "code_len": 1024}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5369, "started": 1791535444.875184}
{"ep": "h2-00591", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.6155, "speedup": 1.6155, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 167.8, "calls": 2, "tokens_in": 5795, "tokens_out": 5126, "ctx_chars": 0, "t_end": 1791535490.3962004, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.6155, "lo95": 1.5968, "mean": 1.6061, "ev_only": 1.6149, "ref_ms": 8.3939, "cand_ms": 5.19583, "n_trials": 186, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 45.76, "compile_s": 1.92, "r_error": "", "src_len": 3076, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 3308, "ts": 1791535351.8803787, "chk": "ok", "code_len": 3076}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4029, "tout": 1818, "ts": 1791535444.5630534, "chk": null, "code_len": 3076}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791535322.586691}
{"ep": "h2-00605", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0519, "speedup": 1.0519, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 133.4, "calls": 2, "tokens_in": 11788, "tokens_out": 3853, "ctx_chars": 6513, "t_end": 1791535490.4044478, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0519, "lo95": 1.0517, "mean": 1.0517, "ev_only": 1.0524, "ref_ms": 15.70698, "cand_ms": 14.93226, "n_trials": 83, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.144, "r_wall_s": 22.6, "compile_s": 2.01, "r_error": "", "src_len": 3897, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1952, "ts": 1791535373.2605402, "chk": "ok", "code_len": 3897}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6926, "tout": 1901, "ts": 1791535467.2001946, "chk": null, "code_len": 3897}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4862, "started": 1791535357.0120938}
{"ep": "h2-00495", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0179, "speedup": 1.0179, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 501.8, "calls": 2, "tokens_in": 15507, "tokens_out": 3406, "ctx_chars": 11823, "t_end": 1791535492.5087903, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0179, "lo95": 1.017, "mean": 1.0184, "ev_only": 1.0183, "ref_ms": 27.56154, "cand_ms": 27.07629, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.012, "r_wall_s": 79.63, "compile_s": 5.61, "r_error": "", "src_len": 2284, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 2143, "ts": 1791535010.783213, "chk": "ok", "code_len": 2284}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8584, "tout": 1263, "ts": 1791535412.859172, "chk": null, "code_len": 2284}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791534990.6720643}
{"ep": "h2-00612", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0009, "speedup": 1.0009, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.2, "calls": 3, "tokens_in": 17520, "tokens_out": 1167, "ctx_chars": 9028, "t_end": 1791535492.8026493, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0009, "lo95": 0.9975, "mean": 1.0025, "ev_only": 1.0024, "ref_ms": 8.05315, "cand_ms": 8.04613, "n_trials": 86, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 34.04, "compile_s": 1.92, "r_error": "", "src_len": 1032, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 550, "ts": 1791535372.5979068, "chk": "ok", "code_len": 1032}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6031, "tout": 65, "ts": 1791535453.1573625, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6120, "tout": 552, "ts": 1791535458.6654377, "chk": null, "code_len": 1032}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5369, "started": 1791535367.5696852}
{"ep": "h2-00630", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0129, "speedup": 1.0129, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 78.0, "calls": 3, "tokens_in": 25004, "tokens_out": 2405, "ctx_chars": 12352, "t_end": 1791535493.6900778, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0129, "lo95": 1.0123, "mean": 1.0126, "ev_only": 1.013, "ref_ms": 64.34895, "cand_ms": 63.5298, "n_trials": 19, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0017, "r_wall_s": 42.82, "compile_s": 1.87, "r_error": "", "src_len": 1930, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1359, "ts": 1791535430.7794833, "chk": "ok", "code_len": 2145}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8799, "tout": 54, "ts": 1791535440.3510928, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8877, "tout": 992, "ts": 1791535450.8233216, "chk": null, "code_len": 1930}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791535416.2495399}
{"ep": "h2-00579", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0827, "speedup": 1.0827, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 233.9, "calls": 4, "tokens_in": 14356, "tokens_out": 3630, "ctx_chars": 0, "t_end": 1791535507.6105063, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0827, "lo95": 1.0823, "mean": 1.0833, "ev_only": 1.0834, "ref_ms": 15.49141, "cand_ms": 14.30819, "n_trials": 82, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1178, "r_wall_s": 64.92, "compile_s": 1.9, "r_error": "", "src_len": 2452, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1902, "tout": 1131, "ts": 1791535285.3780487, "chk": "no_custom_kernel", "code_len": 2284}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 3157, "tout": 1336, "ts": 1791535416.7116182, "chk": "ok", "code_len": 2452}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 4605, "tout": 63, "ts": 1791535431.7660272, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 4692, "tout": 1100, "ts": 1791535442.0117826, "chk": null, "code_len": 2452}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", "no_code", null], "first_tin": 1902, "started": 1791535273.664701}
{"ep": "h2-00634", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0266, "speedup": 1.0266, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 68.7, "calls": 1, "tokens_in": 5832, "tokens_out": 524, "ctx_chars": 9656, "t_end": 1791535508.0043163, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0266, "lo95": 1.026, "mean": 1.0273, "ev_only": 1.0273, "ref_ms": 24.91825, "cand_ms": 24.2727, "n_trials": 53, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0286, "r_wall_s": 62.93, "compile_s": 2.0, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 524, "ts": 1791535444.196885, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791535439.2678177}
{"ep": "h2-00588", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0119, "speedup": 1.0119, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 195.0, "calls": 2, "tokens_in": 15162, "tokens_out": 5309, "ctx_chars": 10823, "t_end": 1791535508.1715333, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0119, "lo95": 1.0112, "mean": 1.012, "ev_only": 1.012, "ref_ms": 30.08418, "cand_ms": 29.72959, "n_trials": 45, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0165, "r_wall_s": 132.83, "compile_s": 3.6, "r_error": "", "src_len": 3719, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 3418, "ts": 1791535342.2970986, "chk": "ok", "code_len": 3719}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8807, "tout": 1891, "ts": 1791535374.6504052, "chk": null, "code_len": 3719}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791535313.174319}
{"ep": "h2-00639", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8024, "speedup": 0.8024, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 49.0, "calls": 3, "tokens_in": 19213, "tokens_out": 1736, "ctx_chars": 8224, "t_end": 1791535516.7265742, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8024, "lo95": 0.8006, "mean": 0.8047, "ev_only": 0.802, "ref_ms": 31.38059, "cand_ms": 39.10609, "n_trials": 37, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 19.7, "compile_s": 1.91, "r_error": "", "src_len": 994, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 1196, "ts": 1791535482.2781603, "chk": "ok", "code_len": 994}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6813, "tout": 55, "ts": 1791535492.3404431, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6892, "tout": 485, "ts": 1791535496.983548, "chk": null, "code_len": 994}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791535467.6868677}
{"ep": "h2-00647", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "forbidden", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 29.1, "calls": 1, "tokens_in": 1964, "tokens_out": 2636, "ctx_chars": 0, "t_end": 1791535516.9570427, "error": "static check failed: line 57: getattr needs a literal, non-private attribute name; line 58: getattr needs a literal, non-private attribute name; line 59: getatt", "f_status": "forbidden", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "triton", "own_frac": null, "r_wall_s": 0.01, "compile_s": null, "r_error": "static check failed: line 57: getattr needs a literal, non-private attribute name; line 58: getattr needs a literal, non-private attribute name; line 59: getatt", "src_len": 2843, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1964, "tout": 2636, "ts": 1791535516.9093642, "chk": null, "code_len": 2843}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1964, "started": 1791535487.8413243}
{"ep": "h2-00615", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7721, "speedup": 0.7721, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 149.5, "calls": 3, "tokens_in": 49911, "tokens_out": 14884, "ctx_chars": 2425, "t_end": 1791535522.6442163, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7721, "lo95": 0.7683, "mean": 0.7764, "ev_only": 0.7723, "ref_ms": 18.91032, "cand_ms": 24.49225, "n_trials": 59, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0297, "r_wall_s": 14.75, "compile_s": 1.99, "r_error": "", "src_len": 17569, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7489, "ts": 1791535431.9414358, "chk": "ok", "code_len": 17569}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19149, "tout": 41, "ts": 1791535450.3944855, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19214, "tout": 7354, "ts": 1791535507.5222194, "chk": null, "code_len": 17569}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791535373.1432114}
{"ep": "h2-00607", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.3219, "speedup": 0.3219, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 162.2, "calls": 3, "tokens_in": 23036, "tokens_out": 6385, "ctx_chars": 6513, "t_end": 1791535524.1000128, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.3219, "lo95": 0.3207, "mean": 0.3219, "ev_only": 0.3205, "ref_ms": 3.25126, "cand_ms": 10.09894, "n_trials": 113, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0256, "r_wall_s": 47.95, "compile_s": 2.2, "r_error": "", "src_len": 6578, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 3211, "ts": 1791535387.3431408, "chk": "ok", "code_len": 6578}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8769, "tout": 28, "ts": 1791535451.3166134, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8821, "tout": 3146, "ts": 1791535476.124039, "chk": null, "code_len": 6578}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791535361.9068265}
{"ep": "h2-00631", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.1615, "speedup": 1.1615, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 108.7, "calls": 3, "tokens_in": 30477, "tokens_out": 6562, "ctx_chars": 11065, "t_end": 1791535524.5364742, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.1615, "lo95": 1.1602, "mean": 1.1537, "ev_only": 1.1738, "ref_ms": 1.79558, "cand_ms": 1.54592, "n_trials": 225, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1355, "r_wall_s": 17.74, "compile_s": 1.96, "r_error": "", "src_len": 2620, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 5161, "ts": 1791535464.8132236, "chk": "ok", "code_len": 2620}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 11900, "tout": 26, "ts": 1791535494.803945, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11950, "tout": 1375, "ts": 1791535506.5489416, "chk": null, "code_len": 2620}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6627, "started": 1791535415.8532104}
{"ep": "h2-00609", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.1178, "speedup": 1.1178, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 162.9, "calls": 2, "tokens_in": 15559, "tokens_out": 3652, "ctx_chars": 11065, "t_end": 1791535525.7116005, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.1178, "lo95": 1.1149, "mean": 1.1139, "ev_only": 1.1197, "ref_ms": 1.80003, "cand_ms": 1.61029, "n_trials": 198, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1724, "r_wall_s": 51.29, "compile_s": 2.01, "r_error": "", "src_len": 2892, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 2193, "ts": 1791535385.095107, "chk": "ok", "code_len": 2969}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8932, "tout": 1459, "ts": 1791535474.3886216, "chk": null, "code_len": 2892}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791535362.7618473}
{"ep": "h2-00601", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.9623, "speedup": 0.9623, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 190.4, "calls": 2, "tokens_in": 8356, "tokens_out": 6467, "ctx_chars": 0, "t_end": 1791535527.7207317, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9623, "lo95": 0.9528, "mean": 0.8525, "ev_only": 0.9653, "ref_ms": 3.30966, "cand_ms": 3.43949, "n_trials": 204, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1001, "r_wall_s": 123.06, "compile_s": 2.96, "r_error": "", "src_len": 5330, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 4015, "ts": 1791535370.5361865, "chk": "ok", "code_len": 5330}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5611, "tout": 2452, "ts": 1791535404.635582, "chk": null, "code_len": 5330}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791535337.3313828}
{"ep": "h2-00625", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4264, "speedup": 1.4264, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 132.9, "calls": 2, "tokens_in": 12999, "tokens_out": 6059, "ctx_chars": 9033, "t_end": 1791535532.6173007, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.4264, "lo95": 1.426, "mean": 1.4262, "ev_only": 1.427, "ref_ms": 15.57076, "cand_ms": 10.91641, "n_trials": 90, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9658, "r_wall_s": 58.59, "compile_s": 1.98, "r_error": "", "src_len": 2632, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 4335, "ts": 1791535440.5141878, "chk": "ok", "code_len": 2632}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7509, "tout": 1724, "ts": 1791535473.6976388, "chk": null, "code_len": 2632}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791535399.7529826}
{"ep": "h2-00638", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0179, "speedup": 1.0179, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 80.1, "calls": 2, "tokens_in": 16861, "tokens_out": 4465, "ctx_chars": 11065, "t_end": 1791535535.6931472, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0179, "lo95": 1.0145, "mean": 1.0634, "ev_only": 1.0207, "ref_ms": 1.80942, "cand_ms": 1.77767, "n_trials": 200, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2784, "r_wall_s": 24.09, "compile_s": 2.05, "r_error": "", "src_len": 1957, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 3495, "ts": 1791535489.0450191, "chk": "ok", "code_len": 1957}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10234, "tout": 970, "ts": 1791535511.5537171, "chk": null, "code_len": 1957}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791535455.624098}
{"ep": "h2-00610", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9229, "speedup": 0.9229, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 179.8, "calls": 3, "tokens_in": 24329, "tokens_out": 2436, "ctx_chars": 11539, "t_end": 1791535543.7020316, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9229, "lo95": 0.9223, "mean": 0.9337, "ev_only": 0.9236, "ref_ms": 7.75674, "cand_ms": 8.40502, "n_trials": 142, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0605, "r_wall_s": 59.97, "compile_s": 1.82, "r_error": "", "src_len": 1816, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1161, "ts": 1791535377.004666, "chk": "ok", "code_len": 1816}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8383, "tout": 426, "ts": 1791535475.263794, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8833, "tout": 849, "ts": 1791535482.9116526, "chk": null, "code_len": 1816}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791535363.9342685}
{"ep": "h2-00586", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0262, "speedup": 1.0262, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 234.6, "calls": 2, "tokens_in": 5486, "tokens_out": 5679, "ctx_chars": 0, "t_end": 1791535544.0076826, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0262, "lo95": 1.0259, "mean": 1.0262, "ev_only": 1.0261, "ref_ms": 30.40027, "cand_ms": 29.62337, "n_trials": 57, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0179, "r_wall_s": 164.88, "compile_s": 3.44, "r_error": "", "src_len": 3593, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 3894, "ts": 1791535344.922581, "chk": "ok", "code_len": 3593}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3861, "tout": 1785, "ts": 1791535378.4092262, "chk": null, "code_len": 3593}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791535309.4419444}
{"ep": "h2-00632", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9938, "speedup": 0.9938, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 127.3, "calls": 3, "tokens_in": 22620, "tokens_out": 3139, "ctx_chars": 10823, "t_end": 1791535544.230095, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9938, "lo95": 0.9936, "mean": 0.9938, "ev_only": 0.994, "ref_ms": 30.24764, "cand_ms": 30.43558, "n_trials": 44, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0362, "r_wall_s": 84.86, "compile_s": 2.33, "r_error": "", "src_len": 3027, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1627, "ts": 1791535431.308206, "chk": "ok", "code_len": 3097}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8092, "tout": 59, "ts": 1791535445.7080708, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8175, "tout": 1453, "ts": 1791535458.9152818, "chk": null, "code_len": 3027}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791535416.9792657}
{"ep": "h2-00590", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.3246, "speedup": 0.3246, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 245.0, "calls": 3, "tokens_in": 22476, "tokens_out": 5834, "ctx_chars": 6513, "t_end": 1791535567.2956357, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.3246, "lo95": 0.3153, "mean": 0.3312, "ev_only": 0.3244, "ref_ms": 3.29171, "cand_ms": 10.14111, "n_trials": 97, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0252, "r_wall_s": 187.38, "compile_s": 3.31, "r_error": "", "src_len": 6024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2931, "ts": 1791535347.3444324, "chk": "ok", "code_len": 6024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8489, "tout": 28, "ts": 1791535357.3973305, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8541, "tout": 2875, "ts": 1791535379.8897378, "chk": null, "code_len": 6024}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791535322.2621863}
{"ep": "h2-00645", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0354, "speedup": 1.0354, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 87.5, "calls": 2, "tokens_in": 12924, "tokens_out": 2168, "ctx_chars": 9274, "t_end": 1791535568.9718146, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0354, "lo95": 1.0338, "mean": 1.0375, "ev_only": 1.041, "ref_ms": 8.30166, "cand_ms": 8.01817, "n_trials": 144, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0054, "r_wall_s": 51.56, "compile_s": 2.03, "r_error": "", "src_len": 1837, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1228, "ts": 1791535492.0198162, "chk": "ok", "code_len": 1837}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7132, "tout": 940, "ts": 1791535517.2572708, "chk": null, "code_len": 1837}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791535481.447232}
{"ep": "h2-00652", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0171, "speedup": 1.0171, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 76.9, "calls": 2, "tokens_in": 9712, "tokens_out": 2389, "ctx_chars": 5026, "t_end": 1791535569.3785212, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0171, "lo95": 1.0163, "mean": 1.0169, "ev_only": 1.0182, "ref_ms": 8.16182, "cand_ms": 8.02487, "n_trials": 179, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0113, "r_wall_s": 46.22, "compile_s": 2.04, "r_error": "", "src_len": 1516, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 1614, "ts": 1791535507.4498472, "chk": "ok", "code_len": 1516}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5719, "tout": 775, "ts": 1791535522.8497317, "chk": null, "code_len": 1516}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791535492.5090346}
{"ep": "h2-00620", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0086, "speedup": 1.0086, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 190.1, "calls": 1, "tokens_in": 6367, "tokens_out": 532, "ctx_chars": 10847, "t_end": 1791535575.396112, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0086, "lo95": 1.0075, "mean": 1.0102, "ev_only": 1.0069, "ref_ms": 7.97166, "cand_ms": 7.90333, "n_trials": 184, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0045, "r_wall_s": 179.97, "compile_s": 3.82, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 532, "ts": 1791535395.3454409, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791535385.2478113}
{"ep": "h2-00485", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.6277, "speedup": 0.6277, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 606.8, "calls": 5, "tokens_in": 57166, "tokens_out": 12081, "ctx_chars": 6513, "t_end": 1791535577.5179837, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.6277, "lo95": 0.6243, "mean": 0.627, "ev_only": 0.6251, "ref_ms": 7.12112, "cand_ms": 11.34445, "n_trials": 129, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0852, "r_wall_s": 39.95, "compile_s": 3.37, "r_error": "", "src_len": 3915, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2718, "ts": 1791534994.2335432, "chk": "incorrect", "code_len": 6212}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8495, "tout": 3384, "ts": 1791535437.646753, "chk": "incorrect", "code_len": 5950}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 12011, "tout": 2338, "ts": 1791535468.3835468, "chk": "incorrect", "code_len": 4531}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 14481, "tout": 1921, "ts": 1791535508.932402, "chk": "ok", "code_len": 3915}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 16534, "tout": 1720, "ts": 1791535537.4601645, "chk": null, "code_len": 3915}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "incorrect", "ok", null], "first_tin": 5645, "started": 1791534970.7582927}
{"ep": "h2-00637", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4123, "speedup": 1.4123, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.5, "calls": 2, "tokens_in": 4997, "tokens_out": 4921, "ctx_chars": 0, "t_end": 1791535578.2522025, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.4123, "lo95": 1.4114, "mean": 1.4087, "ev_only": 1.4147, "ref_ms": 15.548, "cand_ms": 11.00923, "n_trials": 96, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9659, "r_wall_s": 61.04, "compile_s": 2.01, "r_error": "", "src_len": 2375, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 3189, "ts": 1791535481.502752, "chk": "ok", "code_len": 2375}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3430, "tout": 1732, "ts": 1791535516.890346, "chk": null, "code_len": 2375}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1567, "started": 1791535452.724315}
{"ep": "h2-00663", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 58.8, "calls": 1, "tokens_in": 1622, "tokens_out": 827, "ctx_chars": 0, "t_end": 1791535584.5489776, "error": "RuntimeError: Error building extension 'bias_relu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=bias_relu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 50.67, "compile_s": 50.66, "r_error": "RuntimeError: Error building extension 'bias_relu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=bias_relu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "src_len": 1766, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1622, "tout": 827, "ts": 1791535533.7715845, "chk": null, "code_len": 1766}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1622, "started": 1791535525.7118556}
{"ep": "h2-00643", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0069, "speedup": 1.0069, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 105.8, "calls": 2, "tokens_in": 16257, "tokens_out": 2870, "ctx_chars": 12117, "t_end": 1791535585.2824945, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0069, "lo95": 1.0055, "mean": 1.0063, "ev_only": 1.0068, "ref_ms": 42.40292, "cand_ms": 42.11211, "n_trials": 37, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 65.24, "compile_s": 2.04, "r_error": "", "src_len": 2125, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1708, "ts": 1791535494.2600012, "chk": "ok", "code_len": 2125}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8858, "tout": 1162, "ts": 1791535519.3415449, "chk": null, "code_len": 2125}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791535480.1262538}
{"ep": "h2-00604", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4725, "speedup": 1.4725, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 233.4, "calls": 2, "tokens_in": 11228, "tokens_out": 4891, "ctx_chars": 6513, "t_end": 1791535588.257242, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.4725, "lo95": 1.4719, "mean": 1.4674, "ev_only": 1.4735, "ref_ms": 15.46192, "cand_ms": 10.50023, "n_trials": 81, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1668, "r_wall_s": 65.04, "compile_s": 2.2, "r_error": "", "src_len": 3372, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3423, "ts": 1791535385.2754145, "chk": "ok", "code_len": 3372}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6621, "tout": 1468, "ts": 1791535522.4405367, "chk": null, "code_len": 3372}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791535354.8392975}
{"ep": "h2-00606", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4651, "speedup": 1.4651, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 235.1, "calls": 2, "tokens_in": 11221, "tokens_out": 5550, "ctx_chars": 6513, "t_end": 1791535593.2728002, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.4651, "lo95": 1.4642, "mean": 1.4661, "ev_only": 1.4663, "ref_ms": 15.49254, "cand_ms": 10.57445, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1885, "r_wall_s": 69.5, "compile_s": 2.04, "r_error": "", "src_len": 3328, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 4049, "ts": 1791535396.4475617, "chk": "ok", "code_len": 3328}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6614, "tout": 1501, "ts": 1791535522.8425195, "chk": null, "code_len": 3328}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791535358.154315}
{"ep": "h2-00656", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.011, "speedup": 1.011, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 85.9, "calls": 2, "tokens_in": 16212, "tokens_out": 2479, "ctx_chars": 12117, "t_end": 1791535593.920363, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.011, "lo95": 1.0096, "mean": 1.0116, "ev_only": 1.0112, "ref_ms": 42.51798, "cand_ms": 42.05435, "n_trials": 32, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 51.94, "compile_s": 1.91, "r_error": "", "src_len": 1999, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1484, "ts": 1791535520.4793763, "chk": "ok", "code_len": 1999}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8813, "tout": 995, "ts": 1791535541.2950408, "chk": null, "code_len": 1999}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791535508.0045643}
{"ep": "h2-00650", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0496, "speedup": 1.0496, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 110.2, "calls": 2, "tokens_in": 5555, "tokens_out": 4790, "ctx_chars": 0, "t_end": 1791535600.6174877, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0496, "lo95": 1.0479, "mean": 1.049, "ev_only": 1.0497, "ref_ms": 27.76497, "cand_ms": 26.45177, "n_trials": 47, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0166, "r_wall_s": 49.81, "compile_s": 11.77, "r_error": "", "src_len": 3220, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 3173, "ts": 1791535517.7217462, "chk": "ok", "code_len": 3220}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3825, "tout": 1617, "ts": 1791535550.7779214, "chk": null, "code_len": 3220}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791535490.3964148}
{"ep": "h2-00513", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0129, "speedup": 1.0129, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 579.7, "calls": 2, "tokens_in": 16189, "tokens_out": 2549, "ctx_chars": 12117, "t_end": 1791535602.3406649, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0129, "lo95": 1.0112, "mean": 1.0133, "ev_only": 1.0131, "ref_ms": 42.57954, "cand_ms": 42.0367, "n_trials": 35, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 219.96, "compile_s": 4.4, "r_error": "", "src_len": 2022, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1544, "ts": 1791535035.1242292, "chk": "ok", "code_len": 2022}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8790, "tout": 1005, "ts": 1791535381.6242218, "chk": null, "code_len": 2022}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791535022.6619835}
{"ep": "h2-00666", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0527, "speedup": 1.0527, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 71.1, "calls": 3, "tokens_in": 18633, "tokens_out": 3761, "ctx_chars": 6513, "t_end": 1791535606.7911365, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0527, "lo95": 1.0525, "mean": 1.0529, "ev_only": 1.0533, "ref_ms": 15.72437, "cand_ms": 14.93706, "n_trials": 96, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1684, "r_wall_s": 25.93, "compile_s": 2.21, "r_error": "", "src_len": 3808, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1885, "ts": 1791535551.1371937, "chk": "ok", "code_len": 3808}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6859, "tout": 29, "ts": 1791535565.5882509, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6912, "tout": 1847, "ts": 1791535580.6155708, "chk": null, "code_len": 3808}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791535535.693381}
{"ep": "h2-00667", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.849, "speedup": 0.849, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 63.8, "calls": 2, "tokens_in": 15402, "tokens_out": 2415, "ctx_chars": 11175, "t_end": 1791535607.5420637, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.849, "lo95": 0.847, "mean": 0.8478, "ev_only": 0.848, "ref_ms": 14.40844, "cand_ms": 16.97185, "n_trials": 68, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 33.99, "compile_s": 1.88, "r_error": "", "src_len": 2563, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7030, "tout": 1230, "ts": 1791535555.9035516, "chk": "ok", "code_len": 2563}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8372, "tout": 1185, "ts": 1791535573.4541478, "chk": null, "code_len": 2563}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7030, "started": 1791535544.65212}
{"ep": "h2-00654", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0514, "speedup": 1.0514, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 121.0, "calls": 3, "tokens_in": 22858, "tokens_out": 2415, "ctx_chars": 10527, "t_end": 1791535614.6544175, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0514, "lo95": 1.0483, "mean": 1.0638, "ev_only": 1.054, "ref_ms": 19.71315, "cand_ms": 18.74902, "n_trials": 65, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.009, "r_wall_s": 84.21, "compile_s": 2.07, "r_error": "", "src_len": 2171, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1267, "ts": 1791535505.4498067, "chk": "ok", "code_len": 2349}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8040, "tout": 93, "ts": 1791535521.2432442, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8157, "tout": 1055, "ts": 1791535530.1840267, "chk": null, "code_len": 2171}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791535493.6903048}
{"ep": "h2-00642", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9945, "speedup": 0.9945, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 141.0, "calls": 2, "tokens_in": 14697, "tokens_out": 2917, "ctx_chars": 10850, "t_end": 1791535616.5500748, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9945, "lo95": 0.9224, "mean": 0.9688, "ev_only": 0.9497, "ref_ms": 0.16252, "cand_ms": 0.16341, "n_trials": 177, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 98.56, "compile_s": 12.15, "r_error": "", "src_len": 2135, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1791, "ts": 1791535492.3665733, "chk": "ok", "code_len": 2135}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8073, "tout": 1126, "ts": 1791535517.596597, "chk": null, "code_len": 2135}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791535475.5035083}
{"ep": "h2-00498", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.019, "speedup": 1.019, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 621.3, "calls": 2, "tokens_in": 15356, "tokens_out": 3046, "ctx_chars": 11823, "t_end": 1791535619.08359, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.019, "lo95": 1.0181, "mean": 1.0194, "ev_only": 1.0186, "ref_ms": 27.54555, "cand_ms": 27.03115, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0146, "r_wall_s": 246.24, "compile_s": 3.44, "r_error": "", "src_len": 1982, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 1981, "ts": 1791535014.7710493, "chk": "ok", "code_len": 1982}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8433, "tout": 1065, "ts": 1791535372.8150253, "chk": null, "code_len": 1982}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791534997.787282}
{"ep": "h2-00653", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4763, "speedup": 1.4763, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 139.2, "calls": 2, "tokens_in": 11287, "tokens_out": 5316, "ctx_chars": 6513, "t_end": 1791535631.9629161, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.4763, "lo95": 1.4749, "mean": 1.4716, "ev_only": 1.4756, "ref_ms": 15.52344, "cand_ms": 10.51483, "n_trials": 67, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3506, "r_wall_s": 72.11, "compile_s": 2.05, "r_error": "", "src_len": 3532, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3744, "ts": 1791535526.2889593, "chk": "ok", "code_len": 3532}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6680, "tout": 1572, "ts": 1791535559.1389005, "chk": null, "code_len": 3532}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791535492.8028677}
{"ep": "h2-00646", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.9405, "speedup": 2.9405, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 149.6, "calls": 2, "tokens_in": 13988, "tokens_out": 4621, "ctx_chars": 9970, "t_end": 1791535632.3947186, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.9405, "lo95": 2.9386, "mean": 2.9348, "ev_only": 2.9526, "ref_ms": 7.8133, "cand_ms": 2.65713, "n_trials": 186, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 90.13, "compile_s": 2.34, "r_error": "", "src_len": 3025, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 3037, "ts": 1791535508.5492883, "chk": "ok", "code_len": 3025}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7988, "tout": 1584, "ts": 1791535541.3137498, "chk": null, "code_len": 3025}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791535482.75453}
{"ep": "h2-00648", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0063, "speedup": 1.0063, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 143.7, "calls": 2, "tokens_in": 6137, "tokens_out": 6673, "ctx_chars": 0, "t_end": 1791535632.5587835, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0063, "lo95": 1.0042, "mean": 1.0071, "ev_only": 1.008, "ref_ms": 30.2573, "cand_ms": 30.06677, "n_trials": 42, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0208, "r_wall_s": 72.96, "compile_s": 2.0, "r_error": "", "src_len": 4294, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 4343, "ts": 1791535525.8891168, "chk": "ok", "code_len": 4294}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4539, "tout": 2330, "ts": 1791535559.1859303, "chk": null, "code_len": 4294}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791535488.8296082}
{"ep": "h2-00655", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0149, "speedup": 1.0149, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.4, "calls": 2, "tokens_in": 5468, "tokens_out": 4678, "ctx_chars": 0, "t_end": 1791535632.9654047, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0149, "lo95": 1.0131, "mean": 1.0148, "ev_only": 1.0156, "ref_ms": 27.71264, "cand_ms": 27.30504, "n_trials": 32, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0158, "r_wall_s": 49.69, "compile_s": 12.26, "r_error": "", "src_len": 3029, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 3079, "ts": 1791535540.6432238, "chk": "ok", "code_len": 3029}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3738, "tout": 1599, "ts": 1791535583.2460427, "chk": null, "code_len": 3029}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791535507.6107318}
{"ep": "h2-00675", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 56.2, "calls": 1, "tokens_in": 1647, "tokens_out": 901, "ctx_chars": 0, "t_end": 1791535634.4514859, "error": "RuntimeError: Error building extension 'diag_mul_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=diag_mul_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 47.24, "compile_s": 47.23, "r_error": "RuntimeError: Error building extension 'diag_mul_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=diag_mul_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "src_len": 1963, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1647, "tout": 901, "ts": 1791535587.1918447, "chk": null, "code_len": 1963}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1647, "started": 1791535578.252477}
{"ep": "h2-00672", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0068, "speedup": 1.0068, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 68.0, "calls": 1, "tokens_in": 6477, "tokens_out": 502, "ctx_chars": 10909, "t_end": 1791535637.4192464, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0068, "lo95": 1.0064, "mean": 1.008, "ev_only": 1.005, "ref_ms": 8.02415, "cand_ms": 7.96979, "n_trials": 92, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0033, "r_wall_s": 62.65, "compile_s": 2.86, "r_error": "", "src_len": 935, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 502, "ts": 1791535574.6899273, "chk": null, "code_len": 935}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791535569.3787894}
{"ep": "h2-00668", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0025, "speedup": 1.0025, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 96.9, "calls": 2, "tokens_in": 11412, "tokens_out": 1126, "ctx_chars": 9028, "t_end": 1791535640.8832645, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0025, "lo95": 1.0019, "mean": 1.0033, "ev_only": 1.0027, "ref_ms": 8.06089, "cand_ms": 8.04068, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0065, "r_wall_s": 67.69, "compile_s": 3.05, "r_error": "", "src_len": 1046, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 562, "ts": 1791535549.3538272, "chk": "ok", "code_len": 1046}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6043, "tout": 564, "ts": 1791535573.1175175, "chk": null, "code_len": 1046}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791535544.007938}
{"ep": "h2-00657", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.3046, "speedup": 0.3046, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 138.7, "calls": 4, "tokens_in": 43097, "tokens_out": 10944, "ctx_chars": 6513, "t_end": 1791535646.8834982, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.3046, "lo95": 0.3031, "mean": 0.305, "ev_only": 0.3035, "ref_ms": 7.6941, "cand_ms": 25.25759, "n_trials": 58, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0517, "r_wall_s": 15.09, "compile_s": 2.34, "r_error": "", "src_len": 5546, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2784, "ts": 1791535532.4668224, "chk": "incorrect", "code_len": 6362}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8561, "tout": 5733, "ts": 1791535594.225052, "chk": "ok", "code_len": 5546}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 14406, "tout": 55, "ts": 1791535612.595221, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 14485, "tout": 2372, "ts": 1791535631.7498302, "chk": null, "code_len": 5546}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791535508.1717815}
{"ep": "h2-00627", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3301, "speedup": 1.3301, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 245.5, "calls": 1, "tokens_in": 5998, "tokens_out": 768, "ctx_chars": 9970, "t_end": 1791535647.9051323, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3301, "lo95": 1.3297, "mean": 1.3299, "ev_only": 1.332, "ref_ms": 7.80208, "cand_ms": 5.8656, "n_trials": 190, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1109, "r_wall_s": 237.66, "compile_s": 3.09, "r_error": "", "src_len": 1435, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 768, "ts": 1791535409.5524933, "chk": null, "code_len": 1435}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791535402.406601}
{"ep": "h2-00674", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0, "speedup": 1.0, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 75.0, "calls": 2, "tokens_in": 13375, "tokens_out": 972, "ctx_chars": 10847, "t_end": 1791535652.5007832, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0, "lo95": 0.9996, "mean": 0.9999, "ev_only": 1.0007, "ref_ms": 8.02984, "cand_ms": 8.02979, "n_trials": 148, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 25.59, "compile_s": 2.13, "r_error": "", "src_len": 821, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 528, "ts": 1791535582.4504714, "chk": "ok", "code_len": 821}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7006, "tout": 444, "ts": 1791535626.813227, "chk": null, "code_len": 821}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791535577.5185173}
{"ep": "h2-00658", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.338, "speedup": 2.338, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 136.0, "calls": 2, "tokens_in": 15275, "tokens_out": 5959, "ctx_chars": 8529, "t_end": 1791535652.758623, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.338, "lo95": 2.3369, "mean": 2.3363, "ev_only": 2.3451, "ref_ms": 8.50601, "cand_ms": 3.63812, "n_trials": 170, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4201, "r_wall_s": 74.9, "compile_s": 2.71, "r_error": "", "src_len": 3464, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 4173, "ts": 1791535553.3471665, "chk": "ok", "code_len": 3464}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9780, "tout": 1786, "ts": 1791535577.822572, "chk": null, "code_len": 3464}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791535516.726848}
{"ep": "h2-00671", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0295, "speedup": 1.0295, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 85.8, "calls": 3, "tokens_in": 27326, "tokens_out": 4562, "ctx_chars": 11065, "t_end": 1791535654.745314, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0295, "lo95": 1.0291, "mean": 1.0285, "ev_only": 1.0239, "ref_ms": 1.80015, "cand_ms": 1.74854, "n_trials": 210, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2744, "r_wall_s": 16.51, "compile_s": 2.44, "r_error": "", "src_len": 1976, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 3588, "ts": 1791535603.7699769, "chk": "ok", "code_len": 1976}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10327, "tout": 21, "ts": 1791535629.9749084, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10372, "tout": 953, "ts": 1791535638.1198187, "chk": null, "code_len": 1976}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6627, "started": 1791535568.972053}
{"ep": "h2-00669", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9778, "speedup": 0.9778, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.1, "calls": 1, "tokens_in": 7032, "tokens_out": 1137, "ctx_chars": 11539, "t_end": 1791535655.3318667, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9778, "lo95": 0.9773, "mean": 1.0137, "ev_only": 0.9774, "ref_ms": 31.44809, "cand_ms": 32.16179, "n_trials": 39, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0182, "r_wall_s": 97.85, "compile_s": 2.47, "r_error": "", "src_len": 1861, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1137, "ts": 1791535556.3117301, "chk": null, "code_len": 1861}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791535544.230313}
{"ep": "h2-00664", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0633, "speedup": 1.0633, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 128.5, "calls": 3, "tokens_in": 24344, "tokens_out": 3134, "ctx_chars": 10527, "t_end": 1791535656.2704463, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0633, "lo95": 1.0627, "mean": 1.0639, "ev_only": 1.0636, "ref_ms": 19.6531, "cand_ms": 18.48387, "n_trials": 81, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0087, "r_wall_s": 76.78, "compile_s": 2.84, "r_error": "", "src_len": 2168, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 2050, "ts": 1791535546.1404715, "chk": "ok", "code_len": 2168}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8823, "tout": 13, "ts": 1791535570.3960853, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8860, "tout": 1071, "ts": 1791535579.4586895, "chk": null, "code_len": 2168}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791535527.7210128}
{"ep": "h2-00682", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0027, "speedup": 1.0027, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 54.3, "calls": 2, "tokens_in": 11412, "tokens_out": 1126, "ctx_chars": 9028, "t_end": 1791535656.6572394, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0027, "lo95": 1.0011, "mean": 1.0054, "ev_only": 1.0028, "ref_ms": 8.07676, "cand_ms": 8.0552, "n_trials": 176, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 25.44, "compile_s": 2.18, "r_error": "", "src_len": 1046, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 562, "ts": 1791535607.7537346, "chk": "ok", "code_len": 1046}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6043, "tout": 564, "ts": 1791535631.129411, "chk": null, "code_len": 1046}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791535602.3409886}
{"ep": "h2-00617", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.009, "speedup": 2.009, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 282.6, "calls": 2, "tokens_in": 32661, "tokens_out": 24841, "ctx_chars": 2425, "t_end": 1791535658.465866, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.009, "lo95": 1.9955, "mean": 2.0096, "ev_only": 2.0053, "ref_ms": 19.36222, "cand_ms": 9.6378, "n_trials": 71, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1512, "r_wall_s": 85.78, "compile_s": 14.7, "r_error": "", "src_len": 19740, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 15923, "ts": 1791535498.599387, "chk": "ok", "code_len": 19740}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 21111, "tout": 8918, "ts": 1791535572.6520512, "chk": null, "code_len": 19740}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791535375.8374531}
{"ep": "h2-00570", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0559, "speedup": 1.0559, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 403.2, "calls": 3, "tokens_in": 23159, "tokens_out": 2546, "ctx_chars": 10527, "t_end": 1791535659.239602, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0559, "lo95": 1.0552, "mean": 1.0566, "ev_only": 1.056, "ref_ms": 19.67079, "cand_ms": 18.62938, "n_trials": 79, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0054, "r_wall_s": 277.04, "compile_s": 6.6, "r_error": "", "src_len": 2144, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1442, "ts": 1791535283.2353685, "chk": "ok", "code_len": 2144}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8215, "tout": 44, "ts": 1791535373.1583111, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8283, "tout": 1060, "ts": 1791535382.1593392, "chk": null, "code_len": 2144}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791535256.0671914}
{"ep": "h2-00568", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.065, "speedup": 1.065, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 410.6, "calls": 2, "tokens_in": 14733, "tokens_out": 2464, "ctx_chars": 10527, "t_end": 1791535662.5512686, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.065, "lo95": 1.0615, "mean": 1.0618, "ev_only": 1.0634, "ref_ms": 19.83451, "cand_ms": 18.62464, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0085, "r_wall_s": 280.86, "compile_s": 5.29, "r_error": "", "src_len": 2356, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1299, "ts": 1791535281.1995435, "chk": "ok", "code_len": 2356}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8072, "tout": 1165, "ts": 1791535381.6492999, "chk": null, "code_len": 2356}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6661, "started": 1791535251.9972005}
{"ep": "h2-00613", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3852, "speedup": 1.3852, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 292.1, "calls": 2, "tokens_in": 15882, "tokens_out": 27019, "ctx_chars": 6513, "t_end": 1791535664.8156211, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.3852, "lo95": 1.385, "mean": 1.385, "ev_only": 1.3841, "ref_ms": 15.69672, "cand_ms": 11.33194, "n_trials": 87, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.4206, "r_wall_s": 38.45, "compile_s": 11.54, "r_error": "", "src_len": 11213, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 21559, "ts": 1791535572.6195269, "chk": "ok", "code_len": 11267}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 11018, "tout": 5460, "ts": 1791535626.101087, "chk": null, "code_len": 11213}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535372.7212126}
{"ep": "h2-00621", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8032, "speedup": 0.8032, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 284.4, "calls": 2, "tokens_in": 4940, "tokens_out": 1580, "ctx_chars": 0, "t_end": 1791535669.6482494, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8032, "lo95": 0.7989, "mean": 0.8018, "ev_only": 0.8033, "ref_ms": 59.89708, "cand_ms": 74.57463, "n_trials": 18, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.0003, "r_wall_s": 82.08, "compile_s": 49.64, "r_error": "", "src_len": 1409, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1984, "tout": 863, "ts": 1791535394.2489018, "chk": "ok", "code_len": 1409}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2956, "tout": 717, "ts": 1791535587.5305204, "chk": null, "code_len": 1409}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1984, "started": 1791535385.2924993}
{"ep": "h2-00611", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0279, "speedup": 1.0279, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 303.7, "calls": 2, "tokens_in": 12716, "tokens_out": 1585, "ctx_chars": 9656, "t_end": 1791535670.6232016, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0279, "lo95": 1.0264, "mean": 1.0278, "ev_only": 1.0277, "ref_ms": 24.93058, "cand_ms": 24.2536, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0265, "r_wall_s": 117.07, "compile_s": 2.08, "r_error": "", "src_len": 1073, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 936, "ts": 1791535375.7046874, "chk": "ok", "code_len": 1073}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6882, "tout": 649, "ts": 1791535552.5939825, "chk": null, "code_len": 1073}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791535366.930656}
{"ep": "h2-00651", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.02, "speedup": 1.02, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 188.0, "calls": 3, "tokens_in": 9512, "tokens_out": 3587, "ctx_chars": 0, "t_end": 1791535678.446877, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.02, "lo95": 1.0175, "mean": 1.004, "ev_only": 1.0236, "ref_ms": 15.57215, "cand_ms": 15.26659, "n_trials": 97, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1324, "r_wall_s": 125.3, "compile_s": 2.06, "r_error": "", "src_len": 2763, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1902, "tout": 1077, "ts": 1791535501.1258533, "chk": "no_custom_kernel", "code_len": 2168}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 3103, "tout": 1292, "ts": 1791535524.848452, "chk": "ok", "code_len": 2830}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 4507, "tout": 1218, "ts": 1791535552.4616385, "chk": null, "code_len": 2763}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1902, "started": 1791535490.4046423}
{"ep": "h2-00541", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.0927, "speedup": 3.0927, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 526.6, "calls": 3, "tokens_in": 16862, "tokens_out": 13790, "ctx_chars": 0, "t_end": 1791535682.6105728, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 3.0927, "lo95": 3.071, "mean": 3.3562, "ev_only": 3.1015, "ref_ms": 60.86595, "cand_ms": 19.68032, "n_trials": 20, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8686, "r_wall_s": 273.4, "compile_s": 13.77, "r_error": "", "src_len": 6355, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 7287, "ts": 1791535220.1442256, "chk": "no_custom_kernel", "code_len": 6400}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 3371, "ts": 1791535289.874159, "chk": "ok", "code_len": 6355}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9086, "tout": 3132, "ts": 1791535409.1718507, "chk": null, "code_len": 6355}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791535155.97374}
{"ep": "h2-00581", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.994, "speedup": 1.994, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 400.8, "calls": 2, "tokens_in": 30797, "tokens_out": 22266, "ctx_chars": 0, "t_end": 1791535683.7178314, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.994, "lo95": 1.4095, "mean": 1.7304, "ev_only": 1.9955, "ref_ms": 25.84234, "cand_ms": 12.95979, "n_trials": 20, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1612, "r_wall_s": 91.85, "compile_s": 11.93, "r_error": "", "src_len": 19605, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 13323, "ts": 1791535385.1747715, "chk": "ok", "code_len": 19605}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 20163, "tout": 8943, "ts": 1791535591.8021984, "chk": null, "code_len": 19605}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791535282.9076917}
{"ep": "h2-00563", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.4609, "speedup": 0.4609, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 440.5, "calls": 2, "tokens_in": 6716, "tokens_out": 10949, "ctx_chars": 0, "t_end": 1791535683.965662, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.4609, "lo95": 0.4606, "mean": 0.4614, "ev_only": 0.4602, "ref_ms": 7.80703, "cand_ms": 16.93928, "n_trials": 45, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 270.59, "compile_s": 15.48, "r_error": "", "src_len": 5473, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 8254, "ts": 1791535318.5783787, "chk": "ok", "code_len": 5473}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4991, "tout": 2695, "ts": 1791535412.710009, "chk": null, "code_len": 5473}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1725, "started": 1791535243.4325054}
{"ep": "h2-00487", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7169, "speedup": 0.7169, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 709.4, "calls": 2, "tokens_in": 16277, "tokens_out": 4297, "ctx_chars": 12352, "t_end": 1791535684.3583384, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7169, "lo95": 0.7151, "mean": 0.7167, "ev_only": 0.7168, "ref_ms": 64.33919, "cand_ms": 89.74537, "n_trials": 13, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0019, "r_wall_s": 272.27, "compile_s": 3.72, "r_error": "", "src_len": 2124, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 3181, "ts": 1791535005.0205216, "chk": "ok", "code_len": 2124}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8947, "tout": 1116, "ts": 1791535412.0700445, "chk": null, "code_len": 2124}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791534974.9484806}
{"ep": "h2-00685", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 72.6, "calls": 1, "tokens_in": 2157, "tokens_out": 1009, "ctx_chars": 0, "t_end": 1791535687.2827353, "error": "RuntimeError: Error building extension 'touch_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=touch_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 51.86, "compile_s": 51.85, "r_error": "RuntimeError: Error building extension 'touch_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=touch_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "src_len": 1342, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2157, "tout": 1009, "ts": 1791535635.3798318, "chk": null, "code_len": 1342}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2157, "started": 1791535614.6546495}
{"ep": "h2-00705", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0044, "speedup": 1.0044, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 24.8, "calls": 1, "tokens_in": 6477, "tokens_out": 489, "ctx_chars": 10909, "t_end": 1791535687.3161535, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0044, "lo95": 1.004, "mean": 1.0041, "ev_only": 1.005, "ref_ms": 8.01, "cand_ms": 7.97479, "n_trials": 167, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 19.55, "compile_s": 2.08, "r_error": "", "src_len": 900, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 489, "ts": 1791535667.6740935, "chk": null, "code_len": 900}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791535662.5516565}
{"ep": "h2-00628", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0068, "speedup": 1.0068, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 285.8, "calls": 1, "tokens_in": 6367, "tokens_out": 543, "ctx_chars": 10847, "t_end": 1791535691.3461368, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0068, "lo95": 1.0066, "mean": 1.0067, "ev_only": 1.0072, "ref_ms": 7.94201, "cand_ms": 7.8887, "n_trials": 208, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 280.62, "compile_s": 2.08, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 543, "ts": 1791535410.646696, "chk": null, "code_len": 999}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791535405.5285597}
{"ep": "h2-00640", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9618, "speedup": 0.9618, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 219.2, "calls": 2, "tokens_in": 14285, "tokens_out": 12774, "ctx_chars": 6513, "t_end": 1791535691.9473197, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9618, "lo95": 0.9609, "mean": 0.9615, "ev_only": 0.9603, "ref_ms": 15.75081, "cand_ms": 16.37688, "n_trials": 66, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.5322, "r_wall_s": 97.93, "compile_s": 1.93, "r_error": "", "src_len": 7579, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 8903, "ts": 1791535550.0021932, "chk": "ok", "code_len": 7745}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9421, "tout": 3871, "ts": 1791535593.626035, "chk": null, "code_len": 7579}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535472.750439}
{"ep": "h2-00662", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9999, "speedup": 0.9999, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 167.6, "calls": 4, "tokens_in": 27228, "tokens_out": 2340, "ctx_chars": 9033, "t_end": 1791535692.1694417, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9999, "lo95": 0.9987, "mean": 1.0462, "ev_only": 1.0002, "ref_ms": 15.51257, "cand_ms": 15.51463, "n_trials": 72, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 114.77, "compile_s": 3.1, "r_error": "", "src_len": 758, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 831, "ts": 1791535535.1337533, "chk": "ok", "code_len": 732}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6428, "tout": 1080, "ts": 1791535559.1300414, "chk": "ok", "code_len": 758}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7617, "tout": 54, "ts": 1791535573.0698032, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7695, "tout": 375, "ts": 1791535577.0068214, "chk": null, "code_len": 758}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 5488, "started": 1791535524.536699}
{"ep": "h2-00580", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 4.3477, "speedup": 4.3477, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 427.4, "calls": 3, "tokens_in": 14494, "tokens_out": 11434, "ctx_chars": 0, "t_end": 1791535702.2674534, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 4.3477, "lo95": 4.3303, "mean": 4.4086, "ev_only": 4.3731, "ref_ms": 7.86177, "cand_ms": 1.80824, "n_trials": 181, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 86.04, "compile_s": 11.6, "r_error": "", "src_len": 5623, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 5776, "ts": 1791535324.4800806, "chk": "no_custom_kernel", "code_len": 5645}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4940, "tout": 2898, "ts": 1791535373.0943837, "chk": "ok", "code_len": 5623}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7829, "tout": 2760, "ts": 1791535615.4553652, "chk": null, "code_len": 5623}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1725, "started": 1791535274.913098}
{"ep": "h2-00684", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9813, "speedup": 0.9813, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 98.5, "calls": 3, "tokens_in": 18701, "tokens_out": 1538, "ctx_chars": 9033, "t_end": 1791535706.024162, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9813, "lo95": 0.9806, "mean": 1.0305, "ev_only": 0.9818, "ref_ms": 15.45629, "cand_ms": 15.75136, "n_trials": 40, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 65.63, "compile_s": 1.93, "r_error": "", "src_len": 688, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 811, "ts": 1791535617.1075644, "chk": "ok", "code_len": 688}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6408, "tout": 373, "ts": 1791535633.974645, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6805, "tout": 354, "ts": 1791535639.9717083, "chk": null, "code_len": 688}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5488, "started": 1791535607.5422812}
{"ep": "h2-00505", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.4971, "speedup": 0.4971, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 703.9, "calls": 5, "tokens_in": 57562, "tokens_out": 12031, "ctx_chars": 6513, "t_end": 1791535711.0863194, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.4971, "lo95": 0.4306, "mean": 0.5846, "ev_only": 0.4957, "ref_ms": 7.35382, "cand_ms": 14.79254, "n_trials": 80, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0835, "r_wall_s": 67.54, "compile_s": 2.83, "r_error": "", "src_len": 3743, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2712, "ts": 1791535030.2263198, "chk": "incorrect", "code_len": 6178}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8489, "tout": 3656, "ts": 1791535321.2218866, "chk": "incorrect", "code_len": 6278}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 12277, "tout": 2192, "ts": 1791535351.6880343, "chk": "incorrect", "code_len": 4196}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 14601, "tout": 1817, "ts": 1791535381.243355, "chk": "ok", "code_len": 3743}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 16550, "tout": 1654, "ts": 1791535643.440893, "chk": null, "code_len": 3743}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "incorrect", "ok", null], "first_tin": 5645, "started": 1791535007.2161846}
{"ep": "h2-00596", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7675, "speedup": 0.7675, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 381.4, "calls": 3, "tokens_in": 50029, "tokens_out": 14741, "ctx_chars": 2425, "t_end": 1791535711.375026, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7675, "lo95": 0.7595, "mean": 0.6997, "ev_only": 0.7678, "ref_ms": 18.96613, "cand_ms": 24.71031, "n_trials": 52, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0299, "r_wall_s": 15.63, "compile_s": 2.11, "r_error": "", "src_len": 17059, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7553, "ts": 1791535392.0298963, "chk": "ok", "code_len": 17059}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19213, "tout": 31, "ts": 1791535639.7093048, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19268, "tout": 7157, "ts": 1791535695.368305, "chk": null, "code_len": 17059}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791535329.9651668}
{"ep": "h2-00676", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0062, "speedup": 1.0062, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 128.3, "calls": 2, "tokens_in": 13637, "tokens_out": 1094, "ctx_chars": 10909, "t_end": 1791535712.8031743, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0062, "lo95": 1.0042, "mean": 1.0057, "ev_only": 1.0058, "ref_ms": 8.01549, "cand_ms": 7.96649, "n_trials": 156, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0038, "r_wall_s": 58.71, "compile_s": 3.14, "r_error": "", "src_len": 938, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 570, "ts": 1791535589.528394, "chk": "ok", "code_len": 938}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7158, "tout": 524, "ts": 1791535653.9707785, "chk": null, "code_len": 938}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791535584.5492282}
{"ep": "h2-00696", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0118, "speedup": 1.0118, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 66.5, "calls": 1, "tokens_in": 1744, "tokens_out": 3601, "ctx_chars": 0, "t_end": 1791535714.447973, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0118, "lo95": 1.0113, "mean": 1.0116, "ev_only": 1.0116, "ref_ms": 64.44889, "cand_ms": 63.69417, "n_trials": 21, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0014, "r_wall_s": 26.03, "compile_s": 2.78, "r_error": "", "src_len": 2612, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1744, "tout": 3601, "ts": 1791535688.1710315, "chk": null, "code_len": 2612}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1744, "started": 1791535647.9053984}
{"ep": "h2-00689", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.995, "speedup": 1.995, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 89.3, "calls": 1, "tokens_in": 1818, "tokens_out": 2522, "ctx_chars": 0, "t_end": 1791535721.7405446, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.995, "lo95": 1.9937, "mean": 1.9896, "ev_only": 1.9916, "ref_ms": 8.51713, "cand_ms": 4.26934, "n_trials": 172, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.5146, "r_wall_s": 65.24, "compile_s": 48.65, "r_error": "", "src_len": 3781, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1818, "tout": 2522, "ts": 1791535656.377732, "chk": null, "code_len": 3781}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1818, "started": 1791535632.3951628}
{"ep": "h2-00649", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0609, "speedup": 1.0609, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 232.3, "calls": 2, "tokens_in": 12887, "tokens_out": 4533, "ctx_chars": 9033, "t_end": 1791535722.635861, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0609, "lo95": 1.0602, "mean": 1.0611, "ev_only": 1.0611, "ref_ms": 15.46849, "cand_ms": 14.58058, "n_trials": 64, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 162.88, "compile_s": 3.59, "r_error": "", "src_len": 2420, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5490, "tout": 2884, "ts": 1791535516.0197222, "chk": "ok", "code_len": 2420}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7397, "tout": 1649, "ts": 1791535559.4029546, "chk": null, "code_len": 2420}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5490, "started": 1791535490.3144588}
{"ep": "h2-00688", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0053, "speedup": 1.0053, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.0, "calls": 2, "tokens_in": 4070, "tokens_out": 2131, "ctx_chars": 0, "t_end": 1791535723.939908, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0053, "lo95": 1.0047, "mean": 1.0057, "ev_only": 1.0074, "ref_ms": 8.01566, "cand_ms": 7.97364, "n_trials": 205, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 39.43, "compile_s": 1.93, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1473, "ts": 1791535645.3790667, "chk": "ok", "code_len": 1031}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2567, "tout": 658, "ts": 1791535684.4440606, "chk": null, "code_len": 996}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791535631.9631627}
{"ep": "h2-00665", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.1954, "speedup": 1.1954, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 195.4, "calls": 2, "tokens_in": 4944, "tokens_out": 5029, "ctx_chars": 0, "t_end": 1791535728.0312488, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.1954, "lo95": 1.195, "mean": 1.1951, "ev_only": 1.1958, "ref_ms": 15.49938, "cand_ms": 12.96536, "n_trials": 102, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9738, "r_wall_s": 42.49, "compile_s": 1.91, "r_error": "", "src_len": 2300, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 3405, "ts": 1791535563.827412, "chk": "ok", "code_len": 2300}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3377, "tout": 1624, "ts": 1791535684.835011, "chk": null, "code_len": 2300}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1567, "started": 1791535532.617529}
{"ep": "h2-00690", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0507, "speedup": 1.0507, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 95.6, "calls": 3, "tokens_in": 25846, "tokens_out": 4484, "ctx_chars": 10167, "t_end": 1791535728.1642537, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0507, "lo95": 1.0478, "mean": 1.0567, "ev_only": 1.0549, "ref_ms": 8.52033, "cand_ms": 8.10921, "n_trials": 149, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 35.17, "compile_s": 2.42, "r_error": "", "src_len": 1705, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3608, "ts": 1791535675.0385244, "chk": "ok", "code_len": 1705}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9824, "tout": 70, "ts": 1791535685.5856495, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9918, "tout": 806, "ts": 1791535692.8959615, "chk": null, "code_len": 1705}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791535632.5589995}
{"ep": "h2-00644", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.6004, "speedup": 0.6004, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 249.2, "calls": 2, "tokens_in": 7188, "tokens_out": 10701, "ctx_chars": 0, "t_end": 1791535728.9616039, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.6004, "lo95": 0.5979, "mean": 0.6027, "ev_only": 0.6005, "ref_ms": 8.62453, "cand_ms": 14.36395, "n_trials": 112, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.9996, "r_wall_s": 57.57, "compile_s": 19.37, "r_error": "", "src_len": 6045, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 7654, "ts": 1791535548.3850436, "chk": "ok", "code_len": 6045}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5368, "tout": 3047, "ts": 1791535671.3528714, "chk": null, "code_len": 6045}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791535479.7282734}
{"ep": "h2-00587", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9907, "speedup": 0.9907, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 417.8, "calls": 3, "tokens_in": 24788, "tokens_out": 4038, "ctx_chars": 10167, "t_end": 1791535730.927632, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9907, "lo95": 0.986, "mean": 0.9937, "ev_only": 0.9975, "ref_ms": 8.43949, "cand_ms": 8.51852, "n_trials": 140, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0575, "r_wall_s": 333.96, "compile_s": 5.56, "r_error": "", "src_len": 1886, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3080, "ts": 1791535345.1032066, "chk": "ok", "code_len": 1886}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9296, "tout": 68, "ts": 1791535389.2330925, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9388, "tout": 890, "ts": 1791535396.894114, "chk": null, "code_len": 1886}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791535313.1468935}
{"ep": "h2-00659", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.0411, "speedup": 0.0411, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 215.0, "calls": 2, "tokens_in": 16723, "tokens_out": 6492, "ctx_chars": 11539, "t_end": 1791535731.9966047, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.0411, "lo95": 0.0409, "mean": 0.0412, "ev_only": 0.0409, "ref_ms": 7.85524, "cand_ms": 191.21787, "n_trials": 12, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 113.53, "compile_s": 1.97, "r_error": "", "src_len": 4063, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 4413, "ts": 1791535556.4639986, "chk": "ok", "code_len": 4063}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9608, "tout": 2079, "ts": 1791535617.8135881, "chk": null, "code_len": 4063}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7115, "started": 1791535516.957249}
{"ep": "h2-00701", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3346, "speedup": 1.3346, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 76.9, "calls": 1, "tokens_in": 5998, "tokens_out": 761, "ctx_chars": 9970, "t_end": 1791535733.1578074, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.3346, "lo95": 1.3337, "mean": 1.3415, "ev_only": 1.3326, "ref_ms": 7.8095, "cand_ms": 5.8515, "n_trials": 190, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0902, "r_wall_s": 68.94, "compile_s": 2.23, "r_error": "", "src_len": 1376, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 761, "ts": 1791535663.4170327, "chk": null, "code_len": 1376}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791535656.2706769}
{"ep": "h2-00670", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.1432, "speedup": 1.1432, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 166.8, "calls": 2, "tokens_in": 13986, "tokens_out": 11915, "ctx_chars": 6513, "t_end": 1791535734.1151493, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.1432, "lo95": 1.1407, "mean": 1.1441, "ev_only": 1.1452, "ref_ms": 15.89144, "cand_ms": 13.90056, "n_trials": 99, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4865, "r_wall_s": 57.11, "compile_s": 1.97, "r_error": "", "src_len": 7237, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 8316, "ts": 1791535639.5095003, "chk": "ok", "code_len": 7237}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9122, "tout": 3599, "ts": 1791535676.6928403, "chk": null, "code_len": 7237}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535567.2959247}
{"ep": "h2-00692", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0053, "speedup": 1.0053, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 102.1, "calls": 2, "tokens_in": 14943, "tokens_out": 5210, "ctx_chars": 10823, "t_end": 1791535736.5570495, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0053, "lo95": 1.0047, "mean": 1.0058, "ev_only": 1.0055, "ref_ms": 30.24531, "cand_ms": 30.08587, "n_trials": 36, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0155, "r_wall_s": 45.13, "compile_s": 1.98, "r_error": "", "src_len": 3292, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 3524, "ts": 1791535665.5017042, "chk": "ok", "code_len": 3292}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8588, "tout": 1686, "ts": 1791535691.0245376, "chk": null, "code_len": 3292}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791535634.4517496}
{"ep": "h2-00695", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.5573, "speedup": 3.5573, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 96.8, "calls": 2, "tokens_in": 5562, "tokens_out": 6191, "ctx_chars": 0, "t_end": 1791535743.6842256, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 3.5573, "lo95": 3.552, "mean": 3.5517, "ev_only": 3.5546, "ref_ms": 27.71189, "cand_ms": 7.79021, "n_trials": 57, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9348, "r_wall_s": 29.19, "compile_s": 2.19, "r_error": "", "src_len": 2875, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 4539, "ts": 1791535688.7628684, "chk": "ok", "code_len": 2875}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3832, "tout": 1652, "ts": 1791535714.4701328, "chk": null, "code_len": 2875}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791535646.8838713}
{"ep": "h2-00660", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.2326, "speedup": 4.2326, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 222.4, "calls": 3, "tokens_in": 12304, "tokens_out": 8512, "ctx_chars": 0, "t_end": 1791535744.9994268, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 4.2326, "lo95": 4.2132, "mean": 4.0887, "ev_only": 4.262, "ref_ms": 1.81173, "cand_ms": 0.42804, "n_trials": 195, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 64.06, "compile_s": 13.35, "r_error": "", "src_len": 3580, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4394, "ts": 1791535563.7021616, "chk": "no_custom_kernel", "code_len": 3602}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4219, "tout": 2191, "ts": 1791535648.7310581, "chk": "ok", "code_len": 3580}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6326, "tout": 1927, "ts": 1791535680.9175591, "chk": null, "code_len": 3580}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1759, "started": 1791535522.6444552}
{"ep": "h2-00600", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.003, "speedup": 1.003, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 410.1, "calls": 2, "tokens_in": 4174, "tokens_out": 2134, "ctx_chars": 0, "t_end": 1791535746.5459678, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.003, "lo95": 1.0019, "mean": 1.0308, "ev_only": 1.0055, "ref_ms": 7.99933, "cand_ms": 7.97553, "n_trials": 156, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 361.52, "compile_s": 4.5, "r_error": "", "src_len": 906, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1538, "ts": 1791535350.8432932, "chk": "ok", "code_len": 906}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2573, "tout": 596, "ts": 1791535384.9503298, "chk": null, "code_len": 906}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791535336.482569}
{"ep": "h2-00567", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.0568, "speedup": 3.0568, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 497.7, "calls": 3, "tokens_in": 15533, "tokens_out": 11855, "ctx_chars": 0, "t_end": 1791535747.8674073, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 3.0568, "lo95": 3.0451, "mean": 3.0662, "ev_only": 3.0746, "ref_ms": 14.33032, "cand_ms": 4.68806, "n_trials": 66, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 57.96, "compile_s": 13.63, "r_error": "", "src_len": 5428, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6395, "ts": 1791535339.341525, "chk": "no_custom_kernel", "code_len": 5427}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5314, "tout": 2880, "ts": 1791535396.8319354, "chk": "ok", "code_len": 5428}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8036, "tout": 2580, "ts": 1791535689.5958219, "chk": null, "code_len": 5428}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2183, "started": 1791535250.1918037}
{"ep": "h2-00681", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.059, "speedup": 1.059, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 147.5, "calls": 4, "tokens_in": 18863, "tokens_out": 4118, "ctx_chars": 0, "t_end": 1791535748.155144, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.059, "lo95": 1.0587, "mean": 1.0592, "ev_only": 1.0594, "ref_ms": 19.66803, "cand_ms": 18.57213, "n_trials": 69, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0081, "r_wall_s": 41.27, "compile_s": 12.35, "r_error": "", "src_len": 2813, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1956, "tout": 1472, "ts": 1791535614.3122678, "chk": "compile_error", "code_len": 3107}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4667, "tout": 1300, "ts": 1791535674.3892262, "chk": "ok", "code_len": 2813}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6079, "tout": 58, "ts": 1791535696.0848386, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6161, "tout": 1288, "ts": 1791535706.8530166, "chk": null, "code_len": 2813}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 1956, "started": 1791535600.6177223}
{"ep": "h2-00719", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0074, "speedup": 1.0074, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 47.4, "calls": 1, "tokens_in": 6367, "tokens_out": 540, "ctx_chars": 10847, "t_end": 1791535749.6588688, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0074, "lo95": 1.0021, "mean": 1.0059, "ev_only": 1.008, "ref_ms": 8.0743, "cand_ms": 8.0146, "n_trials": 110, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 41.89, "compile_s": 2.0, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 540, "ts": 1791535707.443743, "chk": null, "code_len": 999}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791535702.2676995}
{"ep": "h2-00614", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7892, "speedup": 0.7892, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 378.1, "calls": 4, "tokens_in": 26416, "tokens_out": 1970, "ctx_chars": 8224, "t_end": 1791535750.8677442, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7892, "lo95": 0.6637, "mean": 0.7811, "ev_only": 0.7893, "ref_ms": 77.18947, "cand_ms": 97.80379, "n_trials": 16, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0, "r_wall_s": 55.17, "compile_s": 2.3, "r_error": "", "src_len": 921, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 752, "ts": 1791535381.2534704, "chk": "ok", "code_len": 1057}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6389, "tout": 701, "ts": 1791535679.3125749, "chk": "ok", "code_len": 921}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7199, "tout": 77, "ts": 1791535691.254349, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7300, "tout": 440, "ts": 1791535695.4097202, "chk": null, "code_len": 921}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 5528, "started": 1791535372.7907195}
{"ep": "h2-00622", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3305, "speedup": 1.3305, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 363.7, "calls": 1, "tokens_in": 5998, "tokens_out": 704, "ctx_chars": 9970, "t_end": 1791535750.9928946, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3305, "lo95": 1.33, "mean": 1.3302, "ev_only": 1.3322, "ref_ms": 7.80272, "cand_ms": 5.86438, "n_trials": 119, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0587, "r_wall_s": 356.69, "compile_s": 2.9, "r_error": "", "src_len": 1393, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 704, "ts": 1791535393.6336312, "chk": null, "code_len": 1393}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791535387.2690322}
{"ep": "h2-00619", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8025, "speedup": 0.8025, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 369.1, "calls": 3, "tokens_in": 19059, "tokens_out": 1768, "ctx_chars": 8224, "t_end": 1791535751.6108224, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.8025, "lo95": 0.798, "mean": 0.8018, "ev_only": 0.8024, "ref_ms": 61.20446, "cand_ms": 76.26563, "n_trials": 22, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 38.26, "compile_s": 2.15, "r_error": "", "src_len": 1078, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 987, "ts": 1791535394.256735, "chk": "ok", "code_len": 1078}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6624, "tout": 259, "ts": 1791535708.2106621, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6907, "tout": 522, "ts": 1791535713.3054502, "chk": null, "code_len": 1078}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791535382.4723666}
{"ep": "h2-00697", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0239, "speedup": 1.0239, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 104.1, "calls": 2, "tokens_in": 4851, "tokens_out": 3697, "ctx_chars": 0, "t_end": 1791535756.6472566, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0239, "lo95": 1.0236, "mean": 1.0233, "ev_only": 1.024, "ref_ms": 30.44589, "cand_ms": 29.73503, "n_trials": 52, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0102, "r_wall_s": 58.35, "compile_s": 2.68, "r_error": "", "src_len": 2222, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 2558, "ts": 1791535676.361731, "chk": "ok", "code_len": 2222}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3226, "tout": 1139, "ts": 1791535697.6662743, "chk": null, "code_len": 2222}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791535652.5010285}
{"ep": "h2-00636", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 8.3972, "speedup": 8.3972, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 328.6, "calls": 3, "tokens_in": 15889, "tokens_out": 23561, "ctx_chars": 0, "t_end": 1791535776.4342391, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 8.3972, "lo95": 8.3748, "mean": 8.3872, "ev_only": 8.4131, "ref_ms": 64.4069, "cand_ms": 7.67001, "n_trials": 21, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.2566, "r_wall_s": 40.21, "compile_s": 10.9, "r_error": "", "src_len": 6026, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 17043, "ts": 1791535609.176578, "chk": "no_custom_kernel", "code_len": 6090}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5420, "tout": 3424, "ts": 1791535648.55255, "chk": "ok", "code_len": 6026}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8723, "tout": 3094, "ts": 1791535736.1750374, "chk": null, "code_len": 6026}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1746, "started": 1791535447.8741999}
{"ep": "h2-00724", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7883, "speedup": 0.7883, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 66.1, "calls": 2, "tokens_in": 13821, "tokens_out": 862, "ctx_chars": 10850, "t_end": 1791535780.582702, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.7883, "lo95": 0.7827, "mean": 0.7915, "ev_only": 0.8035, "ref_ms": 0.14897, "cand_ms": 0.18897, "n_trials": 192, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 17.7, "compile_s": 1.92, "r_error": "", "src_len": 794, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6622, "tout": 468, "ts": 1791535719.5283053, "chk": "ok", "code_len": 794}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7199, "tout": 394, "ts": 1791535762.6988842, "chk": null, "code_len": 794}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6622, "started": 1791535714.4482152}
{"ep": "h2-00728", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0082, "speedup": 1.0082, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 54.1, "calls": 3, "tokens_in": 19801, "tokens_out": 2117, "ctx_chars": 9274, "t_end": 1791535782.1632528, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0082, "lo95": 1.0068, "mean": 1.0077, "ev_only": 1.0085, "ref_ms": 8.34152, "cand_ms": 8.27373, "n_trials": 141, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.042, "r_wall_s": 25.15, "compile_s": 1.9, "r_error": "", "src_len": 2021, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1060, "ts": 1791535738.3278499, "chk": "ok", "code_len": 2021}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6962, "tout": 63, "ts": 1791535748.2585168, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7049, "tout": 994, "ts": 1791535756.9233, "chk": null, "code_len": 2021}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535728.0314965}
{"ep": "h2-00698", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9494, "speedup": 0.9494, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 129.7, "calls": 2, "tokens_in": 13877, "tokens_out": 6980, "ctx_chars": 6513, "t_end": 1791535782.4750519, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9494, "lo95": 0.9445, "mean": 0.9613, "ev_only": 0.9523, "ref_ms": 3.26106, "cand_ms": 3.43501, "n_trials": 190, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0932, "r_wall_s": 63.74, "compile_s": 2.26, "r_error": "", "src_len": 5622, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 4417, "ts": 1791535688.7058246, "chk": "ok", "code_len": 5622}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8429, "tout": 2563, "ts": 1791535718.486111, "chk": null, "code_len": 5622}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791535652.7588499}
{"ep": "h2-00721", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.841, "speedup": 0.841, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 72.9, "calls": 1, "tokens_in": 10632, "tokens_out": 7195, "ctx_chars": 0, "t_end": 1791535783.9955597, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.841, "lo95": 0.8397, "mean": 0.8426, "ev_only": 0.841, "ref_ms": 18.92927, "cand_ms": 22.5093, "n_trials": 42, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0225, "r_wall_s": 15.18, "compile_s": 2.01, "r_error": "", "src_len": 16598, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 10632, "tout": 7195, "ts": 1791535768.6593, "chk": null, "code_len": 16598}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 10632, "started": 1791535711.0869465}
{"ep": "h2-00714", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.1166, "speedup": 1.1166, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 98.2, "calls": 2, "tokens_in": 19228, "tokens_out": 7408, "ctx_chars": 11065, "t_end": 1791535785.5139387, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.1166, "lo95": 1.1158, "mean": 1.1193, "ev_only": 1.1284, "ref_ms": 1.79142, "cand_ms": 1.60437, "n_trials": 170, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1931, "r_wall_s": 24.09, "compile_s": 1.9, "r_error": "", "src_len": 2978, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 5862, "ts": 1791535740.6258965, "chk": "ok", "code_len": 2978}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 12601, "tout": 1546, "ts": 1791535761.3195379, "chk": null, "code_len": 2978}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791535687.2830029}
{"ep": "h2-00707", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7323, "speedup": 0.7323, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 117.9, "calls": 2, "tokens_in": 14223, "tokens_out": 1893, "ctx_chars": 10850, "t_end": 1791535787.5051413, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7323, "lo95": 0.7146, "mean": 0.6871, "ev_only": 0.691, "ref_ms": 0.15093, "cand_ms": 0.20611, "n_trials": 161, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 89.65, "compile_s": 2.5, "r_error": "", "src_len": 1185, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1279, "ts": 1791535682.3823469, "chk": "ok", "code_len": 1185}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7599, "tout": 614, "ts": 1791535697.4981642, "chk": null, "code_len": 1185}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791535670.18132}
{"ep": "h2-00717", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0142, "speedup": 1.0142, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.8, "calls": 1, "tokens_in": 1750, "tokens_out": 2652, "ctx_chars": 0, "t_end": 1791535789.7681372, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0142, "lo95": 1.0106, "mean": 1.0154, "ev_only": 1.0147, "ref_ms": 8.1989, "cand_ms": 8.0839, "n_trials": 159, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.009, "r_wall_s": 72.81, "compile_s": 2.13, "r_error": "", "src_len": 1540, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1750, "tout": 2652, "ts": 1791535716.6626256, "chk": null, "code_len": 1540}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1750, "started": 1791535691.947559}
{"ep": "h2-00678", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7589, "speedup": 0.7589, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 207.2, "calls": 2, "tokens_in": 14129, "tokens_out": 1521, "ctx_chars": 10850, "t_end": 1791535795.4231112, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.7589, "lo95": 0.7486, "mean": 0.8004, "ev_only": 0.7629, "ref_ms": 0.15494, "cand_ms": 0.20418, "n_trials": 205, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 97.48, "compile_s": 2.01, "r_error": "", "src_len": 845, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1004, "ts": 1791535597.7010384, "chk": "ok", "code_len": 845}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7505, "tout": 517, "ts": 1791535697.8185353, "chk": null, "code_len": 845}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791535588.2575064}
{"ep": "h2-00686", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 36.7163, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 180.3, "calls": 2, "tokens_in": 6825, "tokens_out": 7142, "ctx_chars": 0, "t_end": 1791535796.8279521, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 36.7163, "lo95": 36.3611, "mean": 36.1185, "ev_only": 37.562, "ref_ms": 19.66482, "cand_ms": 0.53559, "n_trials": 69, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 87.16, "compile_s": 13.32, "r_error": "", "src_len": 4987, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4781, "ts": 1791535665.5030146, "chk": "ok", "code_len": 4987}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4867, "tout": 2361, "ts": 1791535709.4074552, "chk": null, "code_len": 4987}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791535616.5502934}
{"ep": "h2-00735", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 64.3, "calls": 1, "tokens_in": 1764, "tokens_out": 2000, "ctx_chars": 0, "t_end": 1791535800.9012468, "error": "RuntimeError: Error building extension 'fused_gemm_act_gn': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_gemm_act_gn -DTORCH_API_INCLUDE_EXTENSION_H", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 45.34, "compile_s": 45.34, "r_error": "RuntimeError: Error building extension 'fused_gemm_act_gn': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=fused_gemm_act_gn -DTORCH_API_INCLUDE_EXTENSION_H", "src_len": 4013, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1764, "tout": 2000, "ts": 1791535755.5162954, "chk": null, "code_len": 4013}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1764, "started": 1791535736.5572865}
{"ep": "h2-00726", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.6198, "speedup": 0.6198, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.8, "calls": 3, "tokens_in": 18737, "tokens_out": 1545, "ctx_chars": 8224, "t_end": 1791535804.3994527, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.6198, "lo95": 0.5055, "mean": 0.6944, "ev_only": 0.6198, "ref_ms": 33.71071, "cand_ms": 54.39252, "n_trials": 23, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 55.85, "compile_s": 2.26, "r_error": "", "src_len": 1061, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 945, "ts": 1791535733.577875, "chk": "ok", "code_len": 1061}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6562, "tout": 81, "ts": 1791535743.4988525, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6667, "tout": 519, "ts": 1791535748.4910865, "chk": null, "code_len": 1061}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791535722.6361287}
{"ep": "h2-00729", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0611, "speedup": 1.0611, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 78.7, "calls": 2, "tokens_in": 13860, "tokens_out": 3351, "ctx_chars": 10167, "t_end": 1791535806.9113648, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0611, "lo95": 1.0607, "mean": 1.0654, "ev_only": 1.0617, "ref_ms": 8.44481, "cand_ms": 7.95828, "n_trials": 147, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.006, "r_wall_s": 17.82, "compile_s": 1.92, "r_error": "", "src_len": 2375, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 2101, "ts": 1791535745.3291163, "chk": "ok", "code_len": 2375}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7754, "tout": 1250, "ts": 1791535788.930378, "chk": null, "code_len": 2375}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791535728.1644824}
{"ep": "h2-00624", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9938, "speedup": 0.9938, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 408.9, "calls": 3, "tokens_in": 22550, "tokens_out": 3024, "ctx_chars": 10823, "t_end": 1791535807.465932, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9938, "lo95": 0.9936, "mean": 0.9942, "ev_only": 0.9943, "ref_ms": 30.09512, "cand_ms": 30.28154, "n_trials": 44, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0373, "r_wall_s": 53.94, "compile_s": 3.31, "r_error": "", "src_len": 2884, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1611, "ts": 1791535413.9169893, "chk": "ok", "code_len": 2884}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8076, "tout": 21, "ts": 1791535741.322068, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8121, "tout": 1392, "ts": 1791535752.8219273, "chk": null, "code_len": 2884}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791535398.6025355}
{"ep": "h2-00691", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0082, "speedup": 1.0082, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 176.3, "calls": 2, "tokens_in": 16760, "tokens_out": 4388, "ctx_chars": 11065, "t_end": 1791535809.2461956, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0082, "lo95": 1.0062, "mean": 1.0051, "ev_only": 1.0019, "ref_ms": 1.80862, "cand_ms": 1.79394, "n_trials": 194, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2947, "r_wall_s": 84.35, "compile_s": 2.78, "r_error": "", "src_len": 2067, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 3394, "ts": 1791535665.4725926, "chk": "ok", "code_len": 2067}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10133, "tout": 994, "ts": 1791535724.8717225, "chk": null, "code_len": 2067}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791535632.965635}
{"ep": "h2-00616", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0614, "speedup": 1.0614, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 438.5, "calls": 3, "tokens_in": 22860, "tokens_out": 2471, "ctx_chars": 10527, "t_end": 1791535813.8564584, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0614, "lo95": 1.0605, "mean": 1.0625, "ev_only": 1.0599, "ref_ms": 19.7644, "cand_ms": 18.62167, "n_trials": 80, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0084, "r_wall_s": 98.72, "compile_s": 2.14, "r_error": "", "src_len": 2320, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1291, "ts": 1791535387.0055218, "chk": "ok", "code_len": 2320}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8064, "tout": 47, "ts": 1791535705.7070708, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8135, "tout": 1133, "ts": 1791535715.0890496, "chk": null, "code_len": 2320}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791535375.3661056}
{"ep": "h2-00608", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3388, "speedup": 1.3388, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 456.7, "calls": 2, "tokens_in": 16662, "tokens_out": 7040, "ctx_chars": 11539, "t_end": 1791535818.9113245, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.3388, "lo95": 1.3373, "mean": 1.3378, "ev_only": 1.3391, "ref_ms": 31.39001, "cand_ms": 23.44671, "n_trials": 48, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9997, "r_wall_s": 86.32, "compile_s": 1.84, "r_error": "", "src_len": 4095, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 4881, "ts": 1791535404.7584312, "chk": "ok", "code_len": 4095}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9628, "tout": 2159, "ts": 1791535731.7455919, "chk": null, "code_len": 4095}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791535362.217057}
{"ep": "h2-00641", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.8013, "speedup": 1.8013, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 346.9, "calls": 3, "tokens_in": 16305, "tokens_out": 12855, "ctx_chars": 0, "t_end": 1791535820.336627, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.8013, "lo95": 1.8002, "mean": 1.8005, "ev_only": 1.8045, "ref_ms": 7.81644, "cand_ms": 4.33925, "n_trials": 223, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 101.7, "compile_s": 18.65, "r_error": "", "src_len": 5579, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6769, "ts": 1791535537.1611826, "chk": "no_custom_kernel", "code_len": 5732}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5567, "tout": 3294, "ts": 1791535605.7379076, "chk": "ok", "code_len": 5579}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8579, "tout": 2792, "ts": 1791535717.5168543, "chk": null, "code_len": 5579}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791535473.4081976}
{"ep": "h2-00751", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8464, "speedup": 0.8464, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 34.8, "calls": 1, "tokens_in": 7030, "tokens_out": 1221, "ctx_chars": 11175, "t_end": 1791535820.349994, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8464, "lo95": 0.8456, "mean": 0.8462, "ev_only": 0.8462, "ref_ms": 14.24859, "cand_ms": 16.83486, "n_trials": 71, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 23.51, "compile_s": 1.92, "r_error": "", "src_len": 2568, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1221, "ts": 1791535796.734967, "chk": null, "code_len": 2568}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791535785.5141788}
{"ep": "h2-00743", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0053, "speedup": 1.0053, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 74.2, "calls": 3, "tokens_in": 19765, "tokens_out": 2054, "ctx_chars": 9274, "t_end": 1791535825.1430755, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0053, "lo95": 1.0047, "mean": 1.0152, "ev_only": 1.0054, "ref_ms": 8.34195, "cand_ms": 8.29776, "n_trials": 151, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0429, "r_wall_s": 28.81, "compile_s": 1.93, "r_error": "", "src_len": 1950, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1067, "ts": 1791535761.4730535, "chk": "ok", "code_len": 1999}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6969, "tout": 13, "ts": 1791535788.0765605, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7006, "tout": 974, "ts": 1791535796.2484848, "chk": null, "code_len": 1950}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535750.9930975}
{"ep": "h2-00683", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0592, "speedup": 1.0592, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 228.6, "calls": 4, "tokens_in": 18754, "tokens_out": 4113, "ctx_chars": 0, "t_end": 1791535835.4251428, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0592, "lo95": 1.0588, "mean": 1.0599, "ev_only": 1.0598, "ref_ms": 19.68112, "cand_ms": 18.58026, "n_trials": 70, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0102, "r_wall_s": 45.9, "compile_s": 11.22, "r_error": "", "src_len": 2937, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1956, "tout": 1455, "ts": 1791535621.0973754, "chk": "compile_error", "code_len": 2915}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4624, "tout": 1326, "ts": 1791535685.855409, "chk": "ok", "code_len": 2937}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6062, "tout": 26, "ts": 1791535778.8293395, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6112, "tout": 1306, "ts": 1791535789.4698389, "chk": null, "code_len": 2937}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 1956, "started": 1791535607.3201513}
{"ep": "h2-00710", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0295, "speedup": 1.0295, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 156.7, "calls": 2, "tokens_in": 4442, "tokens_out": 2191, "ctx_chars": 0, "t_end": 1791535839.3110712, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0295, "lo95": 1.0283, "mean": 1.0296, "ev_only": 1.0284, "ref_ms": 24.97256, "cand_ms": 24.25746, "n_trials": 60, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0463, "r_wall_s": 90.68, "compile_s": 2.12, "r_error": "", "src_len": 989, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1606, "ts": 1791535698.8338435, "chk": "ok", "code_len": 989}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2774, "tout": 585, "ts": 1791535747.8700771, "chk": null, "code_len": 989}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791535682.6109674}
{"ep": "h2-00722", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.5577, "speedup": 0.5577, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 129.2, "calls": 2, "tokens_in": 12730, "tokens_out": 3183, "ctx_chars": 9033, "t_end": 1791535840.563195, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.5577, "lo95": 0.5575, "mean": 0.5579, "ev_only": 0.5577, "ref_ms": 15.46905, "cand_ms": 27.73683, "n_trials": 47, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 47.15, "compile_s": 1.96, "r_error": "", "src_len": 2812, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1645, "ts": 1791535726.6504362, "chk": "ok", "code_len": 2812}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7242, "tout": 1538, "ts": 1791535793.0928843, "chk": null, "code_len": 2812}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5488, "started": 1791535711.375268}
{"ep": "h2-00742", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0507, "speedup": 1.0507, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 91.2, "calls": 1, "tokens_in": 4862, "tokens_out": 1887, "ctx_chars": 6513, "t_end": 1791535842.093834, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0507, "lo95": 1.0497, "mean": 1.0522, "ev_only": 1.0517, "ref_ms": 15.72053, "cand_ms": 14.96241, "n_trials": 71, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.144, "r_wall_s": 75.5, "compile_s": 1.9, "r_error": "", "src_len": 3808, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 4862, "tout": 1887, "ts": 1791535766.335603, "chk": null, "code_len": 3808}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 4862, "started": 1791535750.8680038}
{"ep": "h2-00680", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.0107, "speedup": 0.0107, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 249.4, "calls": 2, "tokens_in": 16559, "tokens_out": 6094, "ctx_chars": 11539, "t_end": 1791535843.3130202, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.0107, "lo95": 0.0107, "mean": 0.0108, "ev_only": 0.0107, "ref_ms": 7.79332, "cand_ms": 725.09126, "n_trials": 11, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 137.8, "compile_s": 2.22, "r_error": "", "src_len": 3720, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 4171, "ts": 1791535628.450375, "chk": "ok", "code_len": 3720}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9444, "tout": 1923, "ts": 1791535704.5220215, "chk": null, "code_len": 3720}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7115, "started": 1791535593.9205878}
{"ep": "h2-00706", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.5925, "speedup": 0.5925, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 184.6, "calls": 2, "tokens_in": 15357, "tokens_out": 11809, "ctx_chars": 6513, "t_end": 1791535849.4427214, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.5925, "lo95": 0.4528, "mean": 0.6312, "ev_only": 0.5941, "ref_ms": 8.43437, "cand_ms": 14.23557, "n_trials": 90, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3412, "r_wall_s": 79.07, "compile_s": 1.98, "r_error": "", "src_len": 7492, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 8289, "ts": 1791535737.0822015, "chk": "ok", "code_len": 7492}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9710, "tout": 3520, "ts": 1791535770.317833, "chk": null, "code_len": 7492}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791535664.8159688}
{"ep": "h2-00718", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0643, "speedup": 1.0643, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 161.1, "calls": 3, "tokens_in": 25394, "tokens_out": 4457, "ctx_chars": 10167, "t_end": 1791535853.2763414, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0643, "lo95": 1.0636, "mean": 1.0654, "ev_only": 1.0633, "ref_ms": 8.53394, "cand_ms": 8.01838, "n_trials": 127, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0037, "r_wall_s": 82.95, "compile_s": 2.68, "r_error": "", "src_len": 2005, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3377, "ts": 1791535728.9575012, "chk": "ok", "code_len": 2074}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9593, "tout": 80, "ts": 1791535761.8454845, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9697, "tout": 1000, "ts": 1791535770.215605, "chk": null, "code_len": 2005}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791535692.169667}
{"ep": "h2-00704", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.056, "speedup": 1.056, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 197.3, "calls": 2, "tokens_in": 6051, "tokens_out": 6040, "ctx_chars": 0, "t_end": 1791535856.5115628, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.056, "lo95": 1.0443, "mean": 1.0551, "ev_only": 1.0588, "ref_ms": 8.59168, "cand_ms": 8.13574, "n_trials": 148, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.006, "r_wall_s": 105.41, "compile_s": 17.9, "r_error": "", "src_len": 5024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 3568, "ts": 1791535688.113127, "chk": "ok", "code_len": 5024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4416, "tout": 2472, "ts": 1791535751.014033, "chk": null, "code_len": 5024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791535659.239859}
{"ep": "h2-00687", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4939, "speedup": 1.4939, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 238.5, "calls": 2, "tokens_in": 9595, "tokens_out": 18197, "ctx_chars": 0, "t_end": 1791535857.608932, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.4939, "lo95": 1.4937, "mean": 1.5033, "ev_only": 1.4945, "ref_ms": 31.37412, "cand_ms": 21.00174, "n_trials": 36, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0153, "r_wall_s": 64.75, "compile_s": 11.39, "r_error": "", "src_len": 8952, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 13380, "ts": 1791535734.5078318, "chk": "ok", "code_len": 8952}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7517, "tout": 4817, "ts": 1791535792.012808, "chk": null, "code_len": 8952}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791535619.0838866}
{"ep": "h2-00744", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0161, "speedup": 1.0161, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 109.1, "calls": 2, "tokens_in": 10577, "tokens_out": 3566, "ctx_chars": 5026, "t_end": 1791535860.683137, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0161, "lo95": 1.0149, "mean": 1.0182, "ev_only": 1.0165, "ref_ms": 8.25846, "cand_ms": 8.12736, "n_trials": 192, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0113, "r_wall_s": 56.16, "compile_s": 3.32, "r_error": "", "src_len": 2108, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 2479, "ts": 1791535774.0549998, "chk": "ok", "code_len": 2108}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6584, "tout": 1087, "ts": 1791535804.4155998, "chk": null, "code_len": 2108}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791535751.6110668}
{"ep": "h2-00741", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0036, "speedup": 1.0036, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 112.1, "calls": 1, "tokens_in": 6477, "tokens_out": 540, "ctx_chars": 10909, "t_end": 1791535861.7976873, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0036, "lo95": 1.0007, "mean": 1.0076, "ev_only": 1.0052, "ref_ms": 8.12868, "cand_ms": 8.09918, "n_trials": 135, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 106.76, "compile_s": 2.93, "r_error": "", "src_len": 1011, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 540, "ts": 1791535754.949854, "chk": null, "code_len": 1011}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791535749.6592581}
{"ep": "h2-00661", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.5713, "speedup": 1.5713, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 339.8, "calls": 3, "tokens_in": 16255, "tokens_out": 12361, "ctx_chars": 0, "t_end": 1791535863.8689356, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.5713, "lo95": 1.5703, "mean": 1.5713, "ev_only": 1.5694, "ref_ms": 7.76013, "cand_ms": 4.93882, "n_trials": 204, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 119.72, "compile_s": 14.05, "r_error": "", "src_len": 5362, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6404, "ts": 1791535578.8980234, "chk": "no_custom_kernel", "code_len": 5387}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5564, "tout": 3200, "ts": 1791535655.608855, "chk": "ok", "code_len": 5362}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8532, "tout": 2757, "ts": 1791535742.9781141, "chk": null, "code_len": 5362}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791535524.1002681}
{"ep": "h2-00752", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0004, "speedup": 1.0004, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.8, "calls": 3, "tokens_in": 19907, "tokens_out": 2169, "ctx_chars": 9274, "t_end": 1791535865.2905388, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0004, "lo95": 0.9978, "mean": 0.9958, "ev_only": 1.0015, "ref_ms": 8.363, "cand_ms": 8.35945, "n_trials": 162, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0452, "r_wall_s": 48.07, "compile_s": 2.46, "r_error": "", "src_len": 1975, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1122, "ts": 1791535798.3172786, "chk": "ok", "code_len": 2055}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7024, "tout": 45, "ts": 1791535808.6361756, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7093, "tout": 1002, "ts": 1791535817.1050136, "chk": null, "code_len": 1975}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535787.5054083}
{"ep": "h2-00739", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0058, "speedup": 1.0058, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 120.9, "calls": 3, "tokens_in": 20139, "tokens_out": 2747, "ctx_chars": 9266, "t_end": 1791535868.7335174, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0058, "lo95": 1.0052, "mean": 1.0276, "ev_only": 1.0052, "ref_ms": 30.66242, "cand_ms": 30.48561, "n_trials": 44, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.04, "r_wall_s": 68.36, "compile_s": 2.36, "r_error": "", "src_len": 2371, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1491, "ts": 1791535765.6610794, "chk": "ok", "code_len": 2371}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7217, "tout": 67, "ts": 1791535789.7365417, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7308, "tout": 1189, "ts": 1791535799.732048, "chk": null, "code_len": 2371}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791535747.867775}
{"ep": "h2-00736", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9898, "speedup": 0.9898, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 130.7, "calls": 3, "tokens_in": 19785, "tokens_out": 2600, "ctx_chars": 9266, "t_end": 1791535874.3617084, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9898, "lo95": 0.9894, "mean": 0.9898, "ev_only": 0.9895, "ref_ms": 30.42474, "cand_ms": 30.7374, "n_trials": 54, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0376, "r_wall_s": 67.58, "compile_s": 2.12, "r_error": "", "src_len": 2488, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1341, "ts": 1791535755.641884, "chk": "ok", "code_len": 2552}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7067, "tout": 13, "ts": 1791535795.9802816, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7104, "tout": 1246, "ts": 1791535806.372357, "chk": null, "code_len": 2488}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791535743.6844473}
{"ep": "h2-00727", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0125, "speedup": 1.0125, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.7, "calls": 3, "tokens_in": 24743, "tokens_out": 2241, "ctx_chars": 12117, "t_end": 1791535878.612882, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0125, "lo95": 1.0117, "mean": 1.0124, "ev_only": 1.0129, "ref_ms": 42.68361, "cand_ms": 42.15645, "n_trials": 35, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 96.25, "compile_s": 2.06, "r_error": "", "src_len": 2350, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1123, "ts": 1791535749.216871, "chk": "ok", "code_len": 2350}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8632, "tout": 58, "ts": 1791535772.5248997, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8714, "tout": 1060, "ts": 1791535781.5753167, "chk": null, "code_len": 2350}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791535723.9401433}
{"ep": "h2-00679", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7768, "speedup": 0.7768, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 293.4, "calls": 5, "tokens_in": 58786, "tokens_out": 13124, "ctx_chars": 6513, "t_end": 1791535886.6936944, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7768, "lo95": 0.7731, "mean": 0.7298, "ev_only": 0.7782, "ref_ms": 7.11883, "cand_ms": 9.16467, "n_trials": 89, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0007, "r_wall_s": 75.87, "compile_s": 1.87, "r_error": "", "src_len": 3842, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2688, "ts": 1791535615.9110332, "chk": "incorrect", "code_len": 6095}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8465, "tout": 4081, "ts": 1791535665.646817, "chk": "incorrect", "code_len": 5814}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 12678, "tout": 1701, "ts": 1791535750.1902444, "chk": "ok", "code_len": 3842}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 14488, "tout": 2867, "ts": 1791535784.1499026, "chk": "incorrect", "code_len": 5837}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 17510, "tout": 1787, "ts": 1791535810.8029482, "chk": null, "code_len": 3842}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "ok", "incorrect", null], "first_tin": 5645, "started": 1791535593.2730417}
{"ep": "h2-00715", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0062, "speedup": 1.0062, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 199.4, "calls": 3, "tokens_in": 19930, "tokens_out": 2188, "ctx_chars": 9274, "t_end": 1791535886.7303932, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0062, "lo95": 1.006, "mean": 1.0084, "ev_only": 1.0048, "ref_ms": 8.28875, "cand_ms": 8.23777, "n_trials": 150, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0447, "r_wall_s": 54.0, "compile_s": 3.68, "r_error": "", "src_len": 1999, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1134, "ts": 1791535698.159453, "chk": "ok", "code_len": 2064}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7036, "tout": 44, "ts": 1791535823.8403437, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7104, "tout": 1010, "ts": 1791535832.414397, "chk": null, "code_len": 1999}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535687.316519}
{"ep": "h2-00732", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9789, "speedup": 0.9789, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 159.7, "calls": 3, "tokens_in": 24111, "tokens_out": 2286, "ctx_chars": 11539, "t_end": 1791535891.7264795, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9789, "lo95": 0.9773, "mean": 0.9782, "ev_only": 0.9759, "ref_ms": 31.59222, "cand_ms": 32.27366, "n_trials": 43, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0237, "r_wall_s": 119.4, "compile_s": 1.95, "r_error": "", "src_len": 1775, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1277, "ts": 1791535745.726827, "chk": "ok", "code_len": 1775}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8419, "tout": 217, "ts": 1791535764.466299, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8660, "tout": 792, "ts": 1791535771.4896507, "chk": null, "code_len": 1775}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7032, "started": 1791535731.996869}
{"ep": "h2-00677", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0579, "speedup": 1.0579, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 308.8, "calls": 3, "tokens_in": 22948, "tokens_out": 2478, "ctx_chars": 10527, "t_end": 1791535894.0355906, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0579, "lo95": 1.0566, "mean": 1.0627, "ev_only": 1.0589, "ref_ms": 19.76253, "cand_ms": 18.68021, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0085, "r_wall_s": 184.45, "compile_s": 1.89, "r_error": "", "src_len": 2214, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1321, "ts": 1791535597.1188383, "chk": "ok", "code_len": 2410}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8094, "tout": 75, "ts": 1791535694.7024693, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8193, "tout": 1082, "ts": 1791535709.5370266, "chk": null, "code_len": 2214}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791535585.2827337}
{"ep": "h2-00734", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8429, "speedup": 0.8429, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 161.6, "calls": 2, "tokens_in": 15697, "tokens_out": 2985, "ctx_chars": 11175, "t_end": 1791535895.7385776, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8429, "lo95": 0.8421, "mean": 0.8434, "ev_only": 0.8423, "ref_ms": 14.31629, "cand_ms": 16.9855, "n_trials": 55, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 129.71, "compile_s": 2.91, "r_error": "", "src_len": 2325, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1749, "ts": 1791535748.7981737, "chk": "ok", "code_len": 2325}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8665, "tout": 1236, "ts": 1791535765.8822248, "chk": null, "code_len": 2325}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791535734.115411}
{"ep": "h2-00709", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.3566, "speedup": 2.3566, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 218.8, "calls": 3, "tokens_in": 23611, "tokens_out": 5267, "ctx_chars": 8529, "t_end": 1791535897.2083187, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.3566, "lo95": 2.3536, "mean": 2.3564, "ev_only": 2.3566, "ref_ms": 8.54627, "cand_ms": 3.62653, "n_trials": 156, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4167, "r_wall_s": 87.12, "compile_s": 3.49, "r_error": "", "src_len": 3491, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 3426, "ts": 1791535708.7309825, "chk": "ok", "code_len": 3491}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9033, "tout": 26, "ts": 1791535795.463862, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9083, "tout": 1815, "ts": 1791535810.0434487, "chk": null, "code_len": 3491}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5495, "started": 1791535678.4471216}
{"ep": "h2-00764", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0037, "speedup": 1.0037, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.3, "calls": 2, "tokens_in": 11409, "tokens_out": 1107, "ctx_chars": 9028, "t_end": 1791535897.6405172, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0037, "lo95": 1.0032, "mean": 1.0024, "ev_only": 1.004, "ref_ms": 7.99866, "cand_ms": 7.969, "n_trials": 149, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 19.9, "compile_s": 1.9, "r_error": "", "src_len": 1024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 559, "ts": 1791535825.8278947, "chk": "ok", "code_len": 1057}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6040, "tout": 548, "ts": 1791535877.6136787, "chk": null, "code_len": 1024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791535820.3502123}
{"ep": "h2-00693", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4724, "speedup": 1.4724, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 261.0, "calls": 2, "tokens_in": 11230, "tokens_out": 4876, "ctx_chars": 6513, "t_end": 1791535898.3777857, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.4724, "lo95": 1.4718, "mean": 1.4697, "ev_only": 1.4725, "ref_ms": 15.48106, "cand_ms": 10.5145, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.191, "r_wall_s": 190.21, "compile_s": 1.97, "r_error": "", "src_len": 3218, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3429, "ts": 1791535677.9749045, "chk": "ok", "code_len": 3218}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6623, "tout": 1447, "ts": 1791535707.27045, "chk": null, "code_len": 3218}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791535637.4194987}
{"ep": "h2-00759", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0143, "speedup": 1.0143, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 94.6, "calls": 2, "tokens_in": 16644, "tokens_out": 4349, "ctx_chars": 11065, "t_end": 1791535902.032302, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0143, "lo95": 1.0008, "mean": 1.0002, "ev_only": 1.0018, "ref_ms": 1.87276, "cand_ms": 1.8464, "n_trials": 243, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2991, "r_wall_s": 28.0, "compile_s": 1.92, "r_error": "", "src_len": 2175, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 3278, "ts": 1791535839.2652895, "chk": "ok", "code_len": 2175}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10017, "tout": 1071, "ts": 1791535873.7819219, "chk": null, "code_len": 2175}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791535807.46617}
{"ep": "h2-00712", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9959, "speedup": 0.9959, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 218.5, "calls": 3, "tokens_in": 22618, "tokens_out": 3088, "ctx_chars": 10823, "t_end": 1791535902.451035, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9959, "lo95": 0.9955, "mean": 0.9957, "ev_only": 0.996, "ref_ms": 30.11981, "cand_ms": 30.2435, "n_trials": 51, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0346, "r_wall_s": 116.45, "compile_s": 2.14, "r_error": "", "src_len": 2993, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1649, "ts": 1791535698.9647174, "chk": "ok", "code_len": 3135}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8114, "tout": 13, "ts": 1791535773.6278913, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8151, "tout": 1426, "ts": 1791535785.545272, "chk": null, "code_len": 2993}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791535683.9659083}
{"ep": "h2-00569", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.0045, "speedup": 2.0045, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 654.5, "calls": 2, "tokens_in": 30691, "tokens_out": 22528, "ctx_chars": 0, "t_end": 1791535906.553085, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.0045, "lo95": 2.0007, "mean": 2.0014, "ev_only": 2.0019, "ref_ms": 18.72394, "cand_ms": 9.34114, "n_trials": 74, "fast1": true, "flags": [], "backend": "mixed", "own_frac": 0.2856, "r_wall_s": 112.51, "compile_s": 17.17, "r_error": "", "src_len": 19745, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 13670, "ts": 1791535377.4354434, "chk": "ok", "code_len": 19745}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 20057, "tout": 8858, "ts": 1791535793.9593885, "chk": null, "code_len": 19745}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791535252.0427487}
{"ep": "h2-00754", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0436, "speedup": 1.0436, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 114.0, "calls": 2, "tokens_in": 5113, "tokens_out": 4249, "ctx_chars": 0, "t_end": 1791535909.451536, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0436, "lo95": 1.0434, "mean": 1.0436, "ev_only": 1.046, "ref_ms": 7.93094, "cand_ms": 7.59941, "n_trials": 190, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 29.68, "compile_s": 2.77, "r_error": "", "src_len": 2423, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 2859, "ts": 1791535821.6228976, "chk": "ok", "code_len": 2423}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3489, "tout": 1390, "ts": 1791535879.6020932, "chk": null, "code_len": 2423}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791535795.9568045}
{"ep": "h2-00673", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.4506, "speedup": 2.4506, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 334.4, "calls": 2, "tokens_in": 12147, "tokens_out": 20410, "ctx_chars": 0, "t_end": 1791535909.7663481, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.4506, "lo95": 1.8658, "mean": 2.2296, "ev_only": 2.4518, "ref_ms": 7.68981, "cand_ms": 3.13787, "n_trials": 180, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.3539, "r_wall_s": 105.52, "compile_s": 51.77, "r_error": "", "src_len": 11165, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 14693, "ts": 1791535704.780193, "chk": "ok", "code_len": 11236}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9203, "tout": 5717, "ts": 1791535804.22491, "chk": null, "code_len": 11165}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791535575.396356}
{"ep": "h2-00694", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 7.6024, "speedup": 7.6024, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 270.2, "calls": 3, "tokens_in": 16696, "tokens_out": 17448, "ctx_chars": 0, "t_end": 1791535911.0952544, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 7.6024, "lo95": 7.5314, "mean": 8.068, "ev_only": 7.6155, "ref_ms": 34.46052, "cand_ms": 4.53282, "n_trials": 32, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9834, "r_wall_s": 77.84, "compile_s": 13.01, "r_error": "", "src_len": 6354, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 10812, "ts": 1791535737.856965, "chk": "no_custom_kernel", "code_len": 6353}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5683, "tout": 3529, "ts": 1791535781.957909, "chk": "ok", "code_len": 6354}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9047, "tout": 3107, "ts": 1791535833.2335505, "chk": null, "code_len": 6354}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791535640.8835552}
{"ep": "h2-00699", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0647, "speedup": 1.0647, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 257.6, "calls": 2, "tokens_in": 10501, "tokens_out": 2302, "ctx_chars": 6513, "t_end": 1791535912.3695428, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0647, "lo95": 1.0639, "mean": 1.0641, "ev_only": 1.0651, "ref_ms": 15.49401, "cand_ms": 14.55252, "n_trials": 97, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1228, "r_wall_s": 216.63, "compile_s": 2.31, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1179, "ts": 1791535665.605292, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5896, "tout": 1123, "ts": 1791535695.0231903, "chk": null, "code_len": 2422}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791535654.7455785}
{"ep": "h2-00749", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.551, "speedup": 0.551, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 130.5, "calls": 3, "tokens_in": 20641, "tokens_out": 3455, "ctx_chars": 9033, "t_end": 1791535912.9594216, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.551, "lo95": 0.5506, "mean": 0.5579, "ev_only": 0.5513, "ref_ms": 15.51498, "cand_ms": 28.16037, "n_trials": 43, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 54.59, "compile_s": 1.87, "r_error": "", "src_len": 2752, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1928, "ts": 1791535802.3605652, "chk": "ok", "code_len": 2752}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7525, "tout": 79, "ts": 1791535845.918292, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7628, "tout": 1448, "ts": 1791535857.8385284, "chk": null, "code_len": 2752}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5488, "started": 1791535782.4755836}
{"ep": "h2-00773", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 57.0, "calls": 1, "tokens_in": 1622, "tokens_out": 878, "ctx_chars": 0, "t_end": 1791535913.5050404, "error": "RuntimeError: Error building extension 'bias_relu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=bias_relu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 48.13, "compile_s": 48.12, "r_error": "RuntimeError: Error building extension 'bias_relu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=bias_relu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/", "src_len": 1799, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1622, "tout": 878, "ts": 1791535865.3360307, "chk": null, "code_len": 1799}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1622, "started": 1791535857.2397907}
{"ep": "h2-00746", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.1345, "speedup": 1.1345, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 137.7, "calls": 2, "tokens_in": 4814, "tokens_out": 4111, "ctx_chars": 0, "t_end": 1791535914.1473594, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.1345, "lo95": 1.1335, "mean": 1.1335, "ev_only": 1.1355, "ref_ms": 7.98208, "cand_ms": 7.03608, "n_trials": 189, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 78.94, "compile_s": 4.87, "r_error": "", "src_len": 2607, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 2642, "ts": 1791535799.4112303, "chk": "ok", "code_len": 2607}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3311, "tout": 1469, "ts": 1791535835.1333854, "chk": null, "code_len": 2607}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791535776.4345086}
{"ep": "h2-00740", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 36.5348, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 166.1, "calls": 2, "tokens_in": 7254, "tokens_out": 7596, "ctx_chars": 0, "t_end": 1791535914.304282, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 36.5348, "lo95": 35.8702, "mean": 34.9743, "ev_only": 37.6162, "ref_ms": 19.74811, "cand_ms": 0.54053, "n_trials": 59, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 86.74, "compile_s": 12.8, "r_error": "", "src_len": 5597, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4882, "ts": 1791535788.5234456, "chk": "ok", "code_len": 5597}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5296, "tout": 2714, "ts": 1791535827.5158052, "chk": null, "code_len": 5597}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791535748.1554134}
{"ep": "h2-00593", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.573, "speedup": 1.573, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 594.3, "calls": 3, "tokens_in": 16144, "tokens_out": 15177, "ctx_chars": 0, "t_end": 1791535919.4930227, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.573, "lo95": 1.5671, "mean": 1.5711, "ev_only": 1.5747, "ref_ms": 42.39868, "cand_ms": 26.95395, "n_trials": 35, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9999, "r_wall_s": 228.77, "compile_s": 13.44, "r_error": "", "src_len": 5953, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 8897, "ts": 1791535410.708963, "chk": "no_custom_kernel", "code_len": 6010}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5516, "tout": 3431, "ts": 1791535625.6090648, "chk": "ok", "code_len": 5953}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8592, "tout": 2849, "ts": 1791535690.0571253, "chk": null, "code_len": 5953}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2036, "started": 1791535325.2334037}
{"ep": "h2-00748", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7167, "speedup": 0.7167, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 137.4, "calls": 2, "tokens_in": 16356, "tokens_out": 3811, "ctx_chars": 12352, "t_end": 1791535919.5781841, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7167, "lo95": 0.7161, "mean": 0.7167, "ev_only": 0.7168, "ref_ms": 64.34007, "cand_ms": 89.77278, "n_trials": 17, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0019, "r_wall_s": 97.17, "compile_s": 3.24, "r_error": "", "src_len": 2243, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 2627, "ts": 1791535806.1352873, "chk": "ok", "code_len": 2243}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9026, "tout": 1184, "ts": 1791535822.3859515, "chk": null, "code_len": 2243}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791535782.1634994}
{"ep": "h2-00763", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0044, "speedup": 1.0044, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 101.3, "calls": 3, "tokens_in": 23842, "tokens_out": 2572, "ctx_chars": 11823, "t_end": 1791535921.660719, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0044, "lo95": 1.002, "mean": 1.0011, "ev_only": 1.0051, "ref_ms": 27.65882, "cand_ms": 27.53776, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0284, "r_wall_s": 37.14, "compile_s": 3.05, "r_error": "", "src_len": 2224, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1386, "ts": 1791535833.3170772, "chk": "ok", "code_len": 2376}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8419, "tout": 59, "ts": 1791535871.2888353, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8502, "tout": 1127, "ts": 1791535884.4059389, "chk": null, "code_len": 2224}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6921, "started": 1791535820.3368938}
{"ep": "h2-00755", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0007, "speedup": 1.0007, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.4, "calls": 2, "tokens_in": 16252, "tokens_out": 2599, "ctx_chars": 12117, "t_end": 1791535922.2102692, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0007, "lo95": 0.9995, "mean": 1.0008, "ev_only": 1.0004, "ref_ms": 42.31538, "cand_ms": 42.28732, "n_trials": 38, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 71.53, "compile_s": 2.12, "r_error": "", "src_len": 2045, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1576, "ts": 1791535809.6420355, "chk": "ok", "code_len": 2045}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8853, "tout": 1023, "ts": 1791535849.9696746, "chk": null, "code_len": 2045}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791535796.8281984}
{"ep": "h2-00758", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9837, "speedup": 0.9837, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 118.6, "calls": 2, "tokens_in": 11793, "tokens_out": 1494, "ctx_chars": 9033, "t_end": 1791535925.5023324, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9837, "lo95": 0.9812, "mean": 0.9821, "ev_only": 0.9834, "ref_ms": 15.51619, "cand_ms": 15.77301, "n_trials": 42, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0085, "r_wall_s": 80.24, "compile_s": 2.14, "r_error": "", "src_len": 655, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 708, "ts": 1791535814.8018556, "chk": "ok", "code_len": 732}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6305, "tout": 786, "ts": 1791535844.9656441, "chk": null, "code_len": 655}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5488, "started": 1791535806.9116008}
{"ep": "h2-00730", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0095, "speedup": 1.0095, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 204.5, "calls": 2, "tokens_in": 15075, "tokens_out": 5277, "ctx_chars": 10823, "t_end": 1791535933.4999893, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0095, "lo95": 1.0073, "mean": 1.0092, "ev_only": 1.0096, "ref_ms": 30.1269, "cand_ms": 29.84453, "n_trials": 44, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0196, "r_wall_s": 148.7, "compile_s": 2.92, "r_error": "", "src_len": 3509, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 3479, "ts": 1791535759.7485812, "chk": "ok", "code_len": 3509}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8720, "tout": 1798, "ts": 1791535784.351301, "chk": null, "code_len": 3509}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791535728.961845}
{"ep": "h2-00731", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9783, "speedup": 0.9783, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 204.0, "calls": 2, "tokens_in": 14828, "tokens_out": 9127, "ctx_chars": 6513, "t_end": 1791535934.975741, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9783, "lo95": 0.9588, "mean": 0.9291, "ev_only": 0.9756, "ref_ms": 3.44976, "cand_ms": 3.52619, "n_trials": 194, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0958, "r_wall_s": 119.66, "compile_s": 3.04, "r_error": "", "src_len": 7044, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 5714, "ts": 1791535778.4996178, "chk": "ok", "code_len": 7044}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9380, "tout": 3413, "ts": 1791535815.2940936, "chk": null, "code_len": 7044}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791535731.4582293}
{"ep": "h2-00700", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.2483, "speedup": 2.2483, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 280.2, "calls": 3, "tokens_in": 16659, "tokens_out": 13247, "ctx_chars": 0, "t_end": 1791535935.508596, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.2483, "lo95": 2.2469, "mean": 2.2463, "ev_only": 2.2515, "ref_ms": 14.30455, "cand_ms": 6.36248, "n_trials": 99, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 117.28, "compile_s": 14.39, "r_error": "", "src_len": 5881, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6988, "ts": 1791535716.8211422, "chk": "no_custom_kernel", "code_len": 5881}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5690, "tout": 3400, "ts": 1791535781.5450916, "chk": "ok", "code_len": 5881}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8786, "tout": 2859, "ts": 1791535818.1329103, "chk": null, "code_len": 5881}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2183, "started": 1791535655.3321006}
{"ep": "h2-00713", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 37.1256, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.2, "calls": 2, "tokens_in": 6938, "tokens_out": 6862, "ctx_chars": 0, "t_end": 1791535936.5500157, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 37.1256, "lo95": 36.9696, "mean": 36.5039, "ev_only": 38.1548, "ref_ms": 19.71391, "cand_ms": 0.53101, "n_trials": 61, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 101.32, "compile_s": 13.94, "r_error": "", "src_len": 5163, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4383, "ts": 1791535719.8880584, "chk": "ok", "code_len": 5163}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4980, "tout": 2479, "ts": 1791535835.197611, "chk": null, "code_len": 5163}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791535684.358811}
{"ep": "h2-00716", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.6384, "speedup": 0.6384, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 245.6, "calls": 2, "tokens_in": 15438, "tokens_out": 14228, "ctx_chars": 6513, "t_end": 1791535936.9802504, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.6384, "lo95": 0.6358, "mean": 0.6456, "ev_only": 0.6389, "ref_ms": 7.1066, "cand_ms": 11.13124, "n_trials": 125, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3578, "r_wall_s": 41.35, "compile_s": 2.02, "r_error": "", "src_len": 7650, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 10701, "ts": 1791535786.9257464, "chk": "ok", "code_len": 7650}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9791, "tout": 3527, "ts": 1791535895.3747253, "chk": null, "code_len": 7650}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791535691.3463655}
{"ep": "h2-00723", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5541, "speedup": 1.5541, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 224.5, "calls": 2, "tokens_in": 6577, "tokens_out": 6812, "ctx_chars": 0, "t_end": 1791535937.2978582, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.5541, "lo95": 1.5536, "mean": 1.5535, "ev_only": 1.5536, "ref_ms": 31.36677, "cand_ms": 20.18306, "n_trials": 52, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9997, "r_wall_s": 144.5, "compile_s": 3.36, "r_error": "", "src_len": 3463, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 4948, "ts": 1791535758.8795831, "chk": "ok", "code_len": 3463}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4499, "tout": 1864, "ts": 1791535791.7634306, "chk": null, "code_len": 3463}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791535712.8034208}
{"ep": "h2-00737", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.2934, "speedup": 1.2934, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 194.4, "calls": 2, "tokens_in": 14194, "tokens_out": 6766, "ctx_chars": 6513, "t_end": 1791535939.4194615, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.2934, "lo95": 1.2805, "mean": 1.5296, "ev_only": 1.2916, "ref_ms": 7.39569, "cand_ms": 5.71794, "n_trials": 133, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0016, "r_wall_s": 48.8, "compile_s": 2.0, "r_error": "", "src_len": 5168, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 4479, "ts": 1791535806.0666647, "chk": "ok", "code_len": 5216}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8547, "tout": 2287, "ts": 1791535890.37235, "chk": null, "code_len": 5168}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791535744.9996576}
{"ep": "h2-00772", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0637, "speedup": 1.0637, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 87.3, "calls": 2, "tokens_in": 13788, "tokens_out": 3120, "ctx_chars": 10167, "t_end": 1791535940.6218405, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0637, "lo95": 1.0605, "mean": 1.0614, "ev_only": 1.0631, "ref_ms": 8.47023, "cand_ms": 7.96301, "n_trials": 132, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0053, "r_wall_s": 52.5, "compile_s": 1.9, "r_error": "", "src_len": 2132, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 2033, "ts": 1791535870.6794548, "chk": "ok", "code_len": 2132}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7682, "tout": 1087, "ts": 1791535887.9426744, "chk": null, "code_len": 2132}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791535853.276594}
{"ep": "h2-00766", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.7831, "speedup": 2.7831, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 106.4, "calls": 1, "tokens_in": 1818, "tokens_out": 6268, "ctx_chars": 0, "t_end": 1791535941.8113754, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.7831, "lo95": 2.7733, "mean": 2.7834, "ev_only": 2.7865, "ref_ms": 8.64436, "cand_ms": 3.10604, "n_trials": 180, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3017, "r_wall_s": 50.9, "compile_s": 1.98, "r_error": "", "src_len": 2933, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1818, "tout": 6268, "ts": 1791535890.851636, "chk": null, "code_len": 2933}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1818, "started": 1791535835.4253924}
{"ep": "h2-00702", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 6.7235, "speedup": 6.7235, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 285.3, "calls": 3, "tokens_in": 30499, "tokens_out": 26825, "ctx_chars": 8224, "t_end": 1791535941.990279, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 6.7235, "lo95": 6.7081, "mean": 6.8284, "ev_only": 6.7413, "ref_ms": 32.40984, "cand_ms": 4.8204, "n_trials": 52, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9996, "r_wall_s": 20.45, "compile_s": 11.51, "r_error": "", "src_len": 8069, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 18025, "ts": 1791535817.1750822, "chk": "no_custom_kernel", "code_len": 8094}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 10324, "tout": 4667, "ts": 1791535873.9248114, "chk": "ok", "code_len": 8069}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 14665, "tout": 4133, "ts": 1791535921.4991248, "chk": null, "code_len": 8069}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791535656.657531}
{"ep": "h2-00767", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.5495, "speedup": 0.5495, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 106.1, "calls": 2, "tokens_in": 5046, "tokens_out": 4591, "ctx_chars": 0, "t_end": 1791535945.4372556, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.5495, "lo95": 0.5492, "mean": 0.5496, "ev_only": 0.5491, "ref_ms": 15.52011, "cand_ms": 28.24193, "n_trials": 55, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.9868, "r_wall_s": 44.68, "compile_s": 2.95, "r_error": "", "src_len": 2394, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 3042, "ts": 1791535866.2860436, "chk": "ok", "code_len": 2394}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3479, "tout": 1549, "ts": 1791535900.1135619, "chk": null, "code_len": 2394}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1567, "started": 1791535839.3113153}
{"ep": "h2-00778", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0094, "speedup": 1.0094, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.3, "calls": 2, "tokens_in": 4137, "tokens_out": 2025, "ctx_chars": 0, "t_end": 1791535946.613557, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0094, "lo95": 1.0082, "mean": 1.01, "ev_only": 1.0081, "ref_ms": 8.01888, "cand_ms": 7.94439, "n_trials": 162, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0045, "r_wall_s": 44.07, "compile_s": 3.2, "r_error": "", "src_len": 866, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1445, "ts": 1791535878.767236, "chk": "ok", "code_len": 866}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2536, "tout": 580, "ts": 1791535902.2513478, "chk": null, "code_len": 866}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791535865.2907977}
{"ep": "h2-00769", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0278, "speedup": 1.0278, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 109.4, "calls": 1, "tokens_in": 5832, "tokens_out": 539, "ctx_chars": 9656, "t_end": 1791535951.463079, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0278, "lo95": 1.026, "mean": 1.0273, "ev_only": 1.0289, "ref_ms": 24.87552, "cand_ms": 24.20254, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0225, "r_wall_s": 87.03, "compile_s": 1.82, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 539, "ts": 1791535863.4485333, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791535842.0941002}
{"ep": "h2-00703", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 16.9719, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 295.5, "calls": 2, "tokens_in": 7675, "tokens_out": 24351, "ctx_chars": 0, "t_end": 1791535953.9488976, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 16.9719, "lo95": 16.9246, "mean": 16.923, "ev_only": 17.0361, "ref_ms": 64.21093, "cand_ms": 3.78336, "n_trials": 26, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 0.6452, "r_wall_s": 34.42, "compile_s": 11.6, "r_error": "", "src_len": 6304, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 21083, "ts": 1791535856.1884127, "chk": "ok", "code_len": 6346}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5929, "tout": 3268, "ts": 1791535919.2764852, "chk": null, "code_len": 6304}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791535658.4661248}
{"ep": "h2-00761", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.3333, "speedup": 2.3333, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 141.7, "calls": 2, "tokens_in": 14555, "tokens_out": 5222, "ctx_chars": 8529, "t_end": 1791535955.592386, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 2.3333, "lo95": 2.3328, "mean": 2.3312, "ev_only": 2.3397, "ref_ms": 8.49251, "cand_ms": 3.63974, "n_trials": 168, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4231, "r_wall_s": 14.03, "compile_s": 2.12, "r_error": "", "src_len": 3458, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 3453, "ts": 1791535844.771585, "chk": "ok", "code_len": 3458}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9060, "tout": 1769, "ts": 1791535941.521641, "chk": null, "code_len": 3458}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791535813.8567123}
{"ep": "h2-00708", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.5937, "speedup": 3.5937, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 285.1, "calls": 3, "tokens_in": 19683, "tokens_out": 19679, "ctx_chars": 0, "t_end": 1791535955.7728043, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 3.5937, "lo95": 3.5812, "mean": 3.5872, "ev_only": 3.596, "ref_ms": 31.52018, "cand_ms": 8.77092, "n_trials": 35, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9809, "r_wall_s": 60.32, "compile_s": 15.46, "r_error": "", "src_len": 8007, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 10906, "ts": 1791535765.9526227, "chk": "no_custom_kernel", "code_len": 7992}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6634, "tout": 4692, "ts": 1791535822.316971, "chk": "ok", "code_len": 8007}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11083, "tout": 4081, "ts": 1791535895.203454, "chk": null, "code_len": 8007}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791535670.6234734}
{"ep": "h2-00796", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8037, "speedup": 0.8037, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 45.0, "calls": 3, "tokens_in": 18765, "tokens_out": 1539, "ctx_chars": 8224, "t_end": 1791535957.4184136, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8037, "lo95": 0.7982, "mean": 0.806, "ev_only": 0.8042, "ref_ms": 32.0347, "cand_ms": 39.86065, "n_trials": 29, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 18.55, "compile_s": 1.95, "r_error": "", "src_len": 1026, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 960, "ts": 1791535924.047958, "chk": "ok", "code_len": 1026}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6577, "tout": 79, "ts": 1791535934.181745, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6680, "tout": 500, "ts": 1791535938.8488538, "chk": null, "code_len": 1026}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791535912.3698416}
{"ep": "h2-00782", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0103, "speedup": 1.0103, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 72.4, "calls": 3, "tokens_in": 24006, "tokens_out": 3667, "ctx_chars": 10167, "t_end": 1791535959.0572038, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0103, "lo95": 1.0098, "mean": 1.0104, "ev_only": 1.0091, "ref_ms": 8.51729, "cand_ms": 8.43086, "n_trials": 158, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0558, "r_wall_s": 18.44, "compile_s": 2.13, "r_error": "", "src_len": 1845, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 2699, "ts": 1791535914.6658535, "chk": "ok", "code_len": 1845}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8915, "tout": 48, "ts": 1791535932.706252, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8987, "tout": 920, "ts": 1791535940.5261092, "chk": null, "code_len": 1845}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791535886.6939738}
{"ep": "h2-00787", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.007, "speedup": 1.007, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 62.7, "calls": 2, "tokens_in": 16347, "tokens_out": 3573, "ctx_chars": 11823, "t_end": 1791535959.9102201, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.007, "lo95": 1.0063, "mean": 1.0073, "ev_only": 1.0053, "ref_ms": 27.62459, "cand_ms": 27.43392, "n_trials": 51, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0226, "r_wall_s": 21.81, "compile_s": 1.98, "r_error": "", "src_len": 2309, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 2393, "ts": 1791535920.257686, "chk": "ok", "code_len": 2309}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9426, "tout": 1180, "ts": 1791535938.0486379, "chk": null, "code_len": 2309}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6921, "started": 1791535897.2085867}
{"ep": "h2-00756", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0115, "speedup": 1.0115, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 159.8, "calls": 3, "tokens_in": 24816, "tokens_out": 2268, "ctx_chars": 12352, "t_end": 1791535960.7219334, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0115, "lo95": 1.0112, "mean": 1.0117, "ev_only": 1.0116, "ref_ms": 64.22087, "cand_ms": 63.48978, "n_trials": 23, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 26.67, "compile_s": 1.84, "r_error": "", "src_len": 1884, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1279, "ts": 1791535813.0831087, "chk": "ok", "code_len": 2025}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8719, "tout": 26, "ts": 1791535925.5498915, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8769, "tout": 963, "ts": 1791535934.0295508, "chk": null, "code_len": 1884}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791535800.9015179}
{"ep": "h2-00771", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0058, "speedup": 1.0058, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 112.2, "calls": 2, "tokens_in": 4192, "tokens_out": 2110, "ctx_chars": 0, "t_end": 1791535961.6715584, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0058, "lo95": 1.0055, "mean": 1.0325, "ev_only": 1.0057, "ref_ms": 7.96517, "cand_ms": 7.91959, "n_trials": 179, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 83.69, "compile_s": 2.58, "r_error": "", "src_len": 911, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 1461, "ts": 1791535863.297183, "chk": "ok", "code_len": 911}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2591, "tout": 649, "ts": 1791535877.822129, "chk": null, "code_len": 911}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791535849.4430118}
{"ep": "h2-00768", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.019, "speedup": 1.019, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 122.4, "calls": 2, "tokens_in": 5429, "tokens_out": 4434, "ctx_chars": 0, "t_end": 1791535962.9783046, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.019, "lo95": 1.0162, "mean": 1.0177, "ev_only": 1.0187, "ref_ms": 27.55137, "cand_ms": 27.03691, "n_trials": 52, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0154, "r_wall_s": 69.8, "compile_s": 11.91, "r_error": "", "src_len": 2874, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 2909, "ts": 1791535865.232142, "chk": "ok", "code_len": 2874}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3699, "tout": 1525, "ts": 1791535893.0716043, "chk": null, "code_len": 2874}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791535840.5634549}
{"ep": "h2-00725", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.9152, "speedup": 2.9152, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 241.7, "calls": 2, "tokens_in": 14009, "tokens_out": 4865, "ctx_chars": 9970, "t_end": 1791535963.4694269, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.9152, "lo95": 2.9143, "mean": 2.9079, "ev_only": 2.9115, "ref_ms": 7.7974, "cand_ms": 2.67476, "n_trials": 219, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9987, "r_wall_s": 71.27, "compile_s": 2.25, "r_error": "", "src_len": 2760, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 3444, "ts": 1791535751.472314, "chk": "ok", "code_len": 2760}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8009, "tout": 1421, "ts": 1791535891.5658698, "chk": null, "code_len": 2760}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791535721.7408023}
{"ep": "h2-00753", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.5637, "speedup": 0.5637, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 182.2, "calls": 2, "tokens_in": 15766, "tokens_out": 11279, "ctx_chars": 6513, "t_end": 1791535972.0002375, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.5637, "lo95": 0.4001, "mean": 0.4831, "ev_only": 0.5639, "ref_ms": 7.2916, "cand_ms": 12.93455, "n_trials": 63, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.2957, "r_wall_s": 67.09, "compile_s": 2.56, "r_error": "", "src_len": 8364, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 7248, "ts": 1791535850.9548512, "chk": "ok", "code_len": 8364}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10119, "tout": 4031, "ts": 1791535904.8682647, "chk": null, "code_len": 8364}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791535789.7683907}
{"ep": "h2-00779", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0173, "speedup": 1.0173, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 106.4, "calls": 3, "tokens_in": 24912, "tokens_out": 2291, "ctx_chars": 12352, "t_end": 1791535975.1233847, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0173, "lo95": 1.0156, "mean": 1.036, "ev_only": 1.0172, "ref_ms": 64.56333, "cand_ms": 63.46523, "n_trials": 22, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0029, "r_wall_s": 66.48, "compile_s": 1.91, "r_error": "", "src_len": 1858, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1327, "ts": 1791535882.4254513, "chk": "ok", "code_len": 1979}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8767, "tout": 26, "ts": 1791535900.5347888, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8817, "tout": 938, "ts": 1791535908.6002946, "chk": null, "code_len": 1858}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791535868.7337883}
{"ep": "h2-00765", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0128, "speedup": 1.0128, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 153.8, "calls": 2, "tokens_in": 16440, "tokens_out": 2706, "ctx_chars": 12352, "t_end": 1791535978.896565, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0128, "lo95": 1.0121, "mean": 1.0127, "ev_only": 1.0134, "ref_ms": 64.44268, "cand_ms": 63.63112, "n_trials": 20, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 45.21, "compile_s": 1.95, "r_error": "", "src_len": 2023, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1672, "ts": 1791535841.359801, "chk": "ok", "code_len": 2023}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9112, "tout": 1034, "ts": 1791535933.6657846, "chk": null, "code_len": 2023}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7328, "started": 1791535825.1433196}
{"ep": "h2-00760", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0211, "speedup": 1.0211, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 173.3, "calls": 2, "tokens_in": 13424, "tokens_out": 4670, "ctx_chars": 9266, "t_end": 1791535982.5524223, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0211, "lo95": 1.0203, "mean": 1.0209, "ev_only": 1.0206, "ref_ms": 30.37083, "cand_ms": 29.74461, "n_trials": 45, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.02, "r_wall_s": 67.56, "compile_s": 2.04, "r_error": "", "src_len": 3234, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 3019, "ts": 1791535834.676743, "chk": "ok", "code_len": 3234}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7808, "tout": 1651, "ts": 1791535914.5284681, "chk": null, "code_len": 3234}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5616, "started": 1791535809.246446}
{"ep": "h2-00747", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.7002, "speedup": 1.7002, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 210.0, "calls": 2, "tokens_in": 9095, "tokens_out": 8918, "ctx_chars": 0, "t_end": 1791535990.5526943, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.7002, "lo95": 1.6947, "mean": 1.6932, "ev_only": 1.7062, "ref_ms": 3.24797, "cand_ms": 1.9103, "n_trials": 199, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1323, "r_wall_s": 50.6, "compile_s": 11.58, "r_error": "", "src_len": 6915, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 5749, "ts": 1791535828.5952768, "chk": "ok", "code_len": 6915}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6350, "tout": 3169, "ts": 1791535939.92951, "chk": null, "code_len": 6915}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791535780.582991}
{"ep": "h2-00824", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0042, "speedup": 1.0042, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 31.0, "calls": 1, "tokens_in": 6367, "tokens_out": 539, "ctx_chars": 10847, "t_end": 1791535990.959695, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0042, "lo95": 1.0034, "mean": 1.0235, "ev_only": 1.0064, "ref_ms": 7.97621, "cand_ms": 7.94311, "n_trials": 158, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 26.04, "compile_s": 2.59, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 539, "ts": 1791535964.8272119, "chk": null, "code_len": 999}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791535959.91046}
{"ep": "h2-00792", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.3275, "speedup": 2.3275, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.2, "calls": 2, "tokens_in": 13176, "tokens_out": 3922, "ctx_chars": 8529, "t_end": 1791535996.7567832, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.3275, "lo95": 2.3203, "mean": 2.3295, "ev_only": 2.3351, "ref_ms": 8.55042, "cand_ms": 3.67365, "n_trials": 171, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4206, "r_wall_s": 41.59, "compile_s": 2.16, "r_error": "", "src_len": 3597, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2074, "ts": 1791535925.38957, "chk": "ok", "code_len": 3636}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7681, "tout": 1848, "ts": 1791535955.1083019, "chk": null, "code_len": 3597}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791535906.5535214}
{"ep": "h2-00738", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.0582, "speedup": 4.0582, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.2, "calls": 3, "tokens_in": 15889, "tokens_out": 13990, "ctx_chars": 0, "t_end": 1791535998.708745, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 4.0582, "lo95": 3.0435, "mean": 3.8539, "ev_only": 4.0596, "ref_ms": 83.89313, "cand_ms": 20.67255, "n_trials": 18, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8945, "r_wall_s": 54.89, "compile_s": 11.29, "r_error": "", "src_len": 5903, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 8037, "ts": 1791535817.829406, "chk": "no_custom_kernel", "code_len": 5925}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5432, "tout": 3089, "ts": 1791535889.4588654, "chk": "ok", "code_len": 5903}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8471, "tout": 2864, "ts": 1791535943.7662404, "chk": null, "code_len": 5903}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791535746.5462556}
{"ep": "h2-00797", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.3213, "speedup": 0.3213, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 86.7, "calls": 3, "tokens_in": 21758, "tokens_out": 5116, "ctx_chars": 6513, "t_end": 1791535999.6112351, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.3213, "lo95": 0.3199, "mean": 0.3194, "ev_only": 0.3189, "ref_ms": 3.13989, "cand_ms": 9.77328, "n_trials": 86, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0249, "r_wall_s": 35.99, "compile_s": 2.11, "r_error": "", "src_len": 5323, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2556, "ts": 1791535934.0035846, "chk": "ok", "code_len": 5354}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8114, "tout": 60, "ts": 1791535943.808905, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8198, "tout": 2500, "ts": 1791535963.597799, "chk": null, "code_len": 5323}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791535912.9596798}
{"ep": "h2-00770", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0066, "speedup": 1.0066, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 160.2, "calls": 3, "tokens_in": 19895, "tokens_out": 2128, "ctx_chars": 9274, "t_end": 1791536003.5414693, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0066, "lo95": 1.006, "mean": 1.0059, "ev_only": 1.0075, "ref_ms": 8.30859, "cand_ms": 8.25389, "n_trials": 188, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0421, "r_wall_s": 87.41, "compile_s": 2.35, "r_error": "", "src_len": 1901, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5790, "tout": 1101, "ts": 1791535853.3303115, "chk": "ok", "code_len": 1950}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7003, "tout": 75, "ts": 1791535907.9358902, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7102, "tout": 952, "ts": 1791535915.9687881, "chk": null, "code_len": 1901}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5790, "started": 1791535843.3132749}
{"ep": "h2-00711", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0191, "speedup": 1.0191, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 320.3, "calls": 4, "tokens_in": 35102, "tokens_out": 3731, "ctx_chars": 11823, "t_end": 1791536004.0100913, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0191, "lo95": 1.0181, "mean": 1.0193, "ev_only": 1.0194, "ref_ms": 27.64022, "cand_ms": 27.12349, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0115, "r_wall_s": 118.48, "compile_s": 2.94, "r_error": "", "src_len": 2058, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1511, "ts": 1791535698.4116595, "chk": "ok", "code_len": 2801}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8544, "tout": 1133, "ts": 1791535831.7879071, "chk": "ok", "code_len": 2058}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9789, "tout": 35, "ts": 1791535876.2196422, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 9848, "tout": 1052, "ts": 1791535885.3908682, "chk": null, "code_len": 2058}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 6921, "started": 1791535683.718078}
{"ep": "h2-00774", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 37.3773, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 147.3, "calls": 2, "tokens_in": 7321, "tokens_out": 7914, "ctx_chars": 0, "t_end": 1791536004.9459517, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 37.3773, "lo95": 37.2853, "mean": 37.9021, "ev_only": 38.2582, "ref_ms": 19.70626, "cand_ms": 0.52723, "n_trials": 68, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 38.85, "compile_s": 12.99, "r_error": "", "src_len": 5850, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 5037, "ts": 1791535929.0327477, "chk": "ok", "code_len": 5850}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5363, "tout": 2877, "ts": 1791535966.0404127, "chk": null, "code_len": 5850}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791535857.6091936}
{"ep": "h2-00775", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8512, "speedup": 0.8512, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 146.7, "calls": 3, "tokens_in": 19625, "tokens_out": 2785, "ctx_chars": 9033, "t_end": 1791536007.3758814, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.8512, "lo95": 0.8508, "mean": 0.8513, "ev_only": 0.8514, "ref_ms": 15.47334, "cand_ms": 18.17816, "n_trials": 74, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 91.55, "compile_s": 2.31, "r_error": "", "src_len": 2439, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1424, "ts": 1791535874.6676013, "chk": "ok", "code_len": 2439}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7021, "tout": 71, "ts": 1791535904.6482642, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7116, "tout": 1290, "ts": 1791535915.4311872, "chk": null, "code_len": 2439}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5488, "started": 1791535860.6833735}
{"ep": "h2-00806", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3313, "speedup": 1.3313, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 76.5, "calls": 1, "tokens_in": 5998, "tokens_out": 743, "ctx_chars": 9970, "t_end": 1791536009.959371, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.3313, "lo95": 1.3308, "mean": 1.3301, "ev_only": 1.3322, "ref_ms": 7.8158, "cand_ms": 5.87095, "n_trials": 201, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0903, "r_wall_s": 68.89, "compile_s": 2.23, "r_error": "", "src_len": 1435, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 743, "ts": 1791535940.3963413, "chk": null, "code_len": 1435}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791535933.5002313}
{"ep": "h2-00776", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0199, "speedup": 1.0199, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 151.5, "calls": 2, "tokens_in": 5831, "tokens_out": 5380, "ctx_chars": 0, "t_end": 1791536013.274533, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0199, "lo95": 1.0184, "mean": 1.0477, "ev_only": 1.0198, "ref_ms": 27.53012, "cand_ms": 26.99392, "n_trials": 51, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0145, "r_wall_s": 51.57, "compile_s": 12.11, "r_error": "", "src_len": 3589, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 3546, "ts": 1791535892.6879897, "chk": "ok", "code_len": 3636}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4101, "tout": 1834, "ts": 1791535961.6748998, "chk": null, "code_len": 3589}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791535861.7979643}
{"ep": "h2-00745", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8714, "speedup": 0.8714, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 259.7, "calls": 3, "tokens_in": 8282, "tokens_out": 2606, "ctx_chars": 0, "t_end": 1791536016.3952384, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.8714, "lo95": 0.8706, "mean": 0.8715, "ev_only": 0.8712, "ref_ms": 27.492, "cand_ms": 31.54906, "n_trials": 50, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1568, "r_wall_s": 127.17, "compile_s": 2.69, "r_error": "", "src_len": 2217, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1728, "tout": 1356, "ts": 1791535769.244112, "chk": "ok", "code_len": 2247}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3196, "tout": 138, "ts": 1791535879.9585063, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 3358, "tout": 1112, "ts": 1791535889.1196055, "chk": null, "code_len": 2217}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1728, "started": 1791535756.6475}
{"ep": "h2-00793", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0248, "speedup": 1.0248, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 109.1, "calls": 2, "tokens_in": 13344, "tokens_out": 4368, "ctx_chars": 9266, "t_end": 1791536018.5172489, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0248, "lo95": 1.0243, "mean": 1.0253, "ev_only": 1.025, "ref_ms": 30.43615, "cand_ms": 29.69996, "n_trials": 53, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0174, "r_wall_s": 61.17, "compile_s": 1.9, "r_error": "", "src_len": 2928, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 2835, "ts": 1791535933.1596859, "chk": "ok", "code_len": 2928}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7728, "tout": 1533, "ts": 1791535956.9032922, "chk": null, "code_len": 2928}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5616, "started": 1791535909.4517994}
{"ep": "h2-00818", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 68.1, "calls": 1, "tokens_in": 2181, "tokens_out": 1046, "ctx_chars": 0, "t_end": 1791536019.536599, "error": "RuntimeError: Error building extension 'ct3d_copy': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=ct3d_copy -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 52.93, "compile_s": 52.92, "r_error": "RuntimeError: Error building extension 'ct3d_copy': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=ct3d_copy -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "src_len": 1537, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2181, "tout": 1046, "ts": 1791535966.5891082, "chk": null, "code_len": 1537}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2181, "started": 1791535951.463317}
{"ep": "h2-00813", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0269, "speedup": 1.0269, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.9, "calls": 1, "tokens_in": 1668, "tokens_out": 1637, "ctx_chars": 0, "t_end": 1791536022.567274, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0269, "lo95": 1.0259, "mean": 1.0264, "ev_only": 1.0271, "ref_ms": 24.83787, "cand_ms": 24.18827, "n_trials": 55, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0227, "r_wall_s": 66.04, "compile_s": 2.28, "r_error": "", "src_len": 988, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1668, "tout": 1637, "ts": 1791535955.7754757, "chk": null, "code_len": 988}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1668, "started": 1791535940.6220696}
{"ep": "h2-00791", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0166, "speedup": 1.0166, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 122.7, "calls": 3, "tokens_in": 25234, "tokens_out": 2533, "ctx_chars": 12352, "t_end": 1791536025.1719265, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0166, "lo95": 1.0161, "mean": 1.0165, "ev_only": 1.017, "ref_ms": 64.40727, "cand_ms": 63.3552, "n_trials": 19, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.003, "r_wall_s": 54.47, "compile_s": 2.62, "r_error": "", "src_len": 1987, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1478, "ts": 1791535918.7509573, "chk": "ok", "code_len": 2027}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8918, "tout": 46, "ts": 1791535962.354673, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8988, "tout": 1009, "ts": 1791535970.6751559, "chk": null, "code_len": 1987}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7328, "started": 1791535902.451283}
{"ep": "h2-00809", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8462, "speedup": 0.8462, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 88.7, "calls": 2, "tokens_in": 15676, "tokens_out": 2885, "ctx_chars": 11175, "t_end": 1791536025.230001, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8462, "lo95": 0.8452, "mean": 0.8925, "ev_only": 0.8459, "ref_ms": 14.25055, "cand_ms": 16.84115, "n_trials": 40, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 49.57, "compile_s": 2.41, "r_error": "", "src_len": 2318, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1710, "ts": 1791535952.478386, "chk": "ok", "code_len": 2318}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8644, "tout": 1175, "ts": 1791535975.459915, "chk": null, "code_len": 2318}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791535936.5502548}
{"ep": "h2-00808", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0016, "speedup": 1.0016, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 100.9, "calls": 1, "tokens_in": 1596, "tokens_out": 2426, "ctx_chars": 0, "t_end": 1791536036.4179344, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0016, "lo95": 1.0011, "mean": 1.0015, "ev_only": 1.002, "ref_ms": 30.07237, "cand_ms": 30.024, "n_trials": 44, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0316, "r_wall_s": 78.48, "compile_s": 45.78, "r_error": "", "src_len": 4912, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1596, "tout": 2426, "ts": 1791535957.4900532, "chk": null, "code_len": 4912}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1596, "started": 1791535935.5089204}
{"ep": "h2-00822", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0596, "speedup": 1.0596, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 82.4, "calls": 2, "tokens_in": 13786, "tokens_out": 3034, "ctx_chars": 10167, "t_end": 1791536039.8523867, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0596, "lo95": 1.0576, "mean": 1.0594, "ev_only": 1.0608, "ref_ms": 8.54046, "cand_ms": 8.06004, "n_trials": 154, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0053, "r_wall_s": 46.78, "compile_s": 2.07, "r_error": "", "src_len": 2126, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6106, "tout": 1939, "ts": 1791535974.3939877, "chk": "ok", "code_len": 2126}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7680, "tout": 1095, "ts": 1791535992.9778247, "chk": null, "code_len": 2126}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6106, "started": 1791535957.418652}
{"ep": "h2-00803", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0179, "speedup": 1.0179, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 125.3, "calls": 2, "tokens_in": 16602, "tokens_out": 3766, "ctx_chars": 11823, "t_end": 1791536046.970852, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0179, "lo95": 1.0172, "mean": 1.0183, "ev_only": 1.018, "ref_ms": 27.54178, "cand_ms": 27.05869, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0114, "r_wall_s": 50.78, "compile_s": 1.86, "r_error": "", "src_len": 2100, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 2648, "ts": 1791535953.6967258, "chk": "ok", "code_len": 2100}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9681, "tout": 1118, "ts": 1791535995.940088, "chk": null, "code_len": 2100}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6921, "started": 1791535921.6609826}
{"ep": "h2-00827", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0061, "speedup": 1.0061, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 89.7, "calls": 2, "tokens_in": 13692, "tokens_out": 1128, "ctx_chars": 10909, "t_end": 1791536052.680706, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0061, "lo95": 1.0048, "mean": 1.0076, "ev_only": 1.0041, "ref_ms": 8.00422, "cand_ms": 7.95554, "n_trials": 76, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0045, "r_wall_s": 52.38, "compile_s": 2.02, "r_error": "", "src_len": 938, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 625, "ts": 1791535968.8768923, "chk": "ok", "code_len": 938}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7213, "tout": 503, "ts": 1791535999.9624991, "chk": null, "code_len": 938}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791535962.9785492}
{"ep": "h2-00780", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.027, "speedup": 1.027, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 180.4, "calls": 2, "tokens_in": 13526, "tokens_out": 4833, "ctx_chars": 9266, "t_end": 1791536054.7525256, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.027, "lo95": 1.0266, "mean": 1.0266, "ev_only": 1.0267, "ref_ms": 30.64558, "cand_ms": 29.84038, "n_trials": 53, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0154, "r_wall_s": 65.54, "compile_s": 2.89, "r_error": "", "src_len": 3299, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 3068, "ts": 1791535900.015663, "chk": "ok", "code_len": 3299}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7910, "tout": 1765, "ts": 1791535988.3753984, "chk": null, "code_len": 3299}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5616, "started": 1791535874.3619921}
{"ep": "h2-00777", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.7813, "speedup": 0.7813, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 191.0, "calls": 2, "tokens_in": 8440, "tokens_out": 13423, "ctx_chars": 0, "t_end": 1791536054.846621, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.7813, "lo95": 0.7803, "mean": 0.7809, "ev_only": 0.7819, "ref_ms": 15.92975, "cand_ms": 20.38748, "n_trials": 79, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3479, "r_wall_s": 68.26, "compile_s": 2.05, "r_error": "", "src_len": 7091, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2161, "tout": 9814, "ts": 1791535950.1178193, "chk": "ok", "code_len": 7091}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6279, "tout": 3609, "ts": 1791535986.375205, "chk": null, "code_len": 7091}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2161, "started": 1791535863.8691936}
{"ep": "h2-00733", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 3.4513, "speedup": 3.4513, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 324.7, "calls": 3, "tokens_in": 26920, "tokens_out": 17686, "ctx_chars": 8224, "t_end": 1791536057.8628178, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 3.4513, "lo95": 3.4426, "mean": 3.449, "ev_only": 3.455, "ref_ms": 60.14127, "cand_ms": 17.42555, "n_trials": 16, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8566, "r_wall_s": 97.71, "compile_s": 13.58, "r_error": "", "src_len": 6364, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 11261, "ts": 1791535834.5824728, "chk": "no_custom_kernel", "code_len": 6379}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9111, "tout": 3438, "ts": 1791535920.4756687, "chk": "ok", "code_len": 6364}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 12279, "tout": 2987, "ts": 1791535960.1159635, "chk": null, "code_len": 6364}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791535733.1581018}
{"ep": "h2-00810", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0296, "speedup": 1.0296, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 124.6, "calls": 2, "tokens_in": 9293, "tokens_out": 2835, "ctx_chars": 5026, "t_end": 1791536061.5653884, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0296, "lo95": 1.0293, "mean": 1.0321, "ev_only": 1.0279, "ref_ms": 8.13939, "cand_ms": 7.90528, "n_trials": 79, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0042, "r_wall_s": 82.73, "compile_s": 1.96, "r_error": "", "src_len": 1477, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 2001, "ts": 1791535955.160199, "chk": "ok", "code_len": 1477}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5298, "tout": 834, "ts": 1791535978.763816, "chk": null, "code_len": 1477}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791535936.9804966}
{"ep": "h2-00832", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 80.7, "calls": 1, "tokens_in": 2159, "tokens_out": 1961, "ctx_chars": 0, "t_end": 1791536063.2519054, "error": "RuntimeError: Error building extension 'mbconv_ops': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mbconv_ops -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 63.14, "compile_s": 63.12, "r_error": "RuntimeError: Error building extension 'mbconv_ops': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mbconv_ops -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/", "src_len": 4025, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2159, "tout": 1961, "ts": 1791536000.0816903, "chk": null, "code_len": 4025}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2159, "started": 1791535983.1116078}
{"ep": "h2-00795", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.782, "speedup": 0.782, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 153.8, "calls": 3, "tokens_in": 47130, "tokens_out": 14718, "ctx_chars": 0, "t_end": 1791536064.9194539, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.782, "lo95": 0.7788, "mean": 0.7805, "ev_only": 0.7823, "ref_ms": 18.56818, "cand_ms": 23.74426, "n_trials": 49, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0301, "r_wall_s": 13.68, "compile_s": 1.9, "r_error": "", "src_len": 17176, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10632, "tout": 7475, "ts": 1791535971.5722508, "chk": "ok", "code_len": 17176}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18219, "tout": 36, "ts": 1791535984.865896, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 18279, "tout": 7207, "ts": 1791536051.0854023, "chk": null, "code_len": 17176}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 10632, "started": 1791535911.0955248}
{"ep": "h2-00840", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0186, "speedup": 1.0186, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 65.5, "calls": 2, "tokens_in": 10564, "tokens_out": 3521, "ctx_chars": 5026, "t_end": 1791536070.4575121, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0186, "lo95": 1.0181, "mean": 1.0186, "ev_only": 1.0189, "ref_ms": 8.13921, "cand_ms": 7.99062, "n_trials": 136, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0114, "r_wall_s": 21.82, "compile_s": 1.82, "r_error": "", "src_len": 2049, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 2466, "ts": 1791536025.7367113, "chk": "ok", "code_len": 2049}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6571, "tout": 1055, "ts": 1791536048.4918015, "chk": null, "code_len": 2049}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791536004.9461977}
{"ep": "h2-00802", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4375, "speedup": 1.4375, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.9, "calls": 2, "tokens_in": 5042, "tokens_out": 4518, "ctx_chars": 0, "t_end": 1791536074.4572241, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.4375, "lo95": 1.4369, "mean": 1.4369, "ev_only": 1.4387, "ref_ms": 15.4688, "cand_ms": 10.76056, "n_trials": 105, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9648, "r_wall_s": 101.18, "compile_s": 2.64, "r_error": "", "src_len": 2478, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 2945, "ts": 1791535945.6996775, "chk": "ok", "code_len": 2478}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3475, "tout": 1573, "ts": 1791535972.9571311, "chk": null, "code_len": 2478}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1567, "started": 1791535919.5784333}
{"ep": "h2-00762", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3799, "speedup": 1.3799, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 257.3, "calls": 2, "tokens_in": 16853, "tokens_out": 8018, "ctx_chars": 11539, "t_end": 1791536076.2448542, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.3799, "lo95": 1.3796, "mean": 1.3797, "ev_only": 1.3803, "ref_ms": 31.39405, "cand_ms": 22.75021, "n_trials": 50, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9601, "r_wall_s": 77.74, "compile_s": 1.95, "r_error": "", "src_len": 4009, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 5958, "ts": 1791535889.2580676, "chk": "ok", "code_len": 4009}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9819, "tout": 2060, "ts": 1791535997.5338962, "chk": null, "code_len": 4009}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791535818.9115965}
{"ep": "h2-00834", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0053, "speedup": 1.0053, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 87.4, "calls": 2, "tokens_in": 11765, "tokens_out": 1497, "ctx_chars": 9028, "t_end": 1791536078.3350036, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0053, "lo95": 1.0043, "mean": 1.0057, "ev_only": 1.0044, "ref_ms": 8.07442, "cand_ms": 8.03174, "n_trials": 126, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0071, "r_wall_s": 49.18, "compile_s": 2.05, "r_error": "", "src_len": 998, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 911, "ts": 1791535999.9013457, "chk": "ok", "code_len": 998}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6394, "tout": 586, "ts": 1791536029.0893974, "chk": null, "code_len": 998}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791535990.959948}
{"ep": "h2-00757", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9994, "speedup": 0.9994, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 274.4, "calls": 4, "tokens_in": 23377, "tokens_out": 6476, "ctx_chars": 0, "t_end": 1791536078.7823627, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9994, "lo95": 0.999, "mean": 1.022, "ev_only": 0.9989, "ref_ms": 15.74787, "cand_ms": 15.75795, "n_trials": 70, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.1465, "r_wall_s": 82.41, "compile_s": 17.97, "r_error": "", "src_len": 2841, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 1928, "ts": 1791535821.9365637, "chk": "compile_error", "code_len": 3822}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5344, "tout": 1680, "ts": 1791535894.1198905, "chk": "incorrect", "code_len": 3715}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 7156, "tout": 1453, "ts": 1791535964.4666195, "chk": "ok", "code_len": 2841}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 8718, "tout": 1415, "ts": 1791535996.2106166, "chk": null, "code_len": 2841}], "n_turns": 4, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "incorrect", "ok", null], "first_tin": 2159, "started": 1791535804.3996832}
{"ep": "h2-00849", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0071, "speedup": 1.0071, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 58.5, "calls": 1, "tokens_in": 6367, "tokens_out": 538, "ctx_chars": 10847, "t_end": 1791536083.7526143, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0071, "lo95": 1.0031, "mean": 1.0087, "ev_only": 1.0074, "ref_ms": 8.04047, "cand_ms": 7.98387, "n_trials": 219, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 52.99, "compile_s": 2.35, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 538, "ts": 1791536030.6968045, "chk": null, "code_len": 999}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791536025.230197}
{"ep": "h2-00836", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.6706, "speedup": 2.6706, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.0, "calls": 2, "tokens_in": 14488, "tokens_out": 5028, "ctx_chars": 8529, "t_end": 1791536090.7162595, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 2.6706, "lo95": 2.6699, "mean": 2.6707, "ev_only": 2.6671, "ref_ms": 8.4962, "cand_ms": 3.18138, "n_trials": 185, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3383, "r_wall_s": 19.49, "compile_s": 2.18, "r_error": "", "src_len": 3151, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 3386, "ts": 1791536028.3240392, "chk": "ok", "code_len": 3151}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8993, "tout": 1642, "ts": 1791536071.1924734, "chk": null, "code_len": 3151}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791535998.7090046}
{"ep": "h2-00781", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0058, "speedup": 1.0058, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 212.6, "calls": 2, "tokens_in": 11680, "tokens_out": 1363, "ctx_chars": 9028, "t_end": 1791536091.234955, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0058, "lo95": 1.0051, "mean": 1.0304, "ev_only": 1.0071, "ref_ms": 7.99394, "cand_ms": 7.94813, "n_trials": 142, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0038, "r_wall_s": 65.78, "compile_s": 2.12, "r_error": "", "src_len": 965, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 826, "ts": 1791535885.8758533, "chk": "ok", "code_len": 965}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6309, "tout": 537, "ts": 1791536025.1487687, "chk": null, "code_len": 965}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791535878.613141}
{"ep": "h2-00799", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7911, "speedup": 0.7911, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 178.0, "calls": 3, "tokens_in": 50114, "tokens_out": 15034, "ctx_chars": 2425, "t_end": 1791536092.1507127, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.7911, "lo95": 0.7886, "mean": 0.7895, "ev_only": 0.7914, "ref_ms": 18.59565, "cand_ms": 23.50602, "n_trials": 55, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0301, "r_wall_s": 16.28, "compile_s": 2.13, "r_error": "", "src_len": 17657, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7586, "ts": 1791535974.4738357, "chk": "ok", "code_len": 17657}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19246, "tout": 50, "ts": 1791536014.3111567, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19320, "tout": 7398, "ts": 1791536075.707302, "chk": null, "code_len": 17657}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791535914.1476326}
{"ep": "h2-00847", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0063, "speedup": 1.0063, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 70.6, "calls": 1, "tokens_in": 6477, "tokens_out": 517, "ctx_chars": 10909, "t_end": 1791536093.1481848, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0063, "lo95": 1.005, "mean": 1.0061, "ev_only": 1.005, "ref_ms": 8.02299, "cand_ms": 7.97239, "n_trials": 75, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 64.87, "compile_s": 2.33, "r_error": "", "src_len": 960, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 517, "ts": 1791536027.9767892, "chk": null, "code_len": 960}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791536022.5675275}
{"ep": "h2-00839", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.4348, "speedup": 2.4348, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.1, "calls": 2, "tokens_in": 13087, "tokens_out": 4277, "ctx_chars": 8529, "t_end": 1791536094.0660734, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 2.4348, "lo95": 2.4342, "mean": 2.4648, "ev_only": 2.4306, "ref_ms": 8.49822, "cand_ms": 3.49027, "n_trials": 192, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4019, "r_wall_s": 24.0, "compile_s": 2.13, "r_error": "", "src_len": 2870, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5497, "tout": 2694, "ts": 1791536026.7090538, "chk": "ok", "code_len": 2870}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7590, "tout": 1583, "ts": 1791536070.024054, "chk": null, "code_len": 2870}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5497, "started": 1791536004.010508}
{"ep": "h2-00783", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3114, "speedup": 1.3114, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 208.1, "calls": 2, "tokens_in": 30597, "tokens_out": 15794, "ctx_chars": 2425, "t_end": 1791536094.7921157, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.3114, "lo95": 1.2951, "mean": 1.3019, "ev_only": 1.3117, "ref_ms": 19.23782, "cand_ms": 14.66969, "n_trials": 81, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0999, "r_wall_s": 75.52, "compile_s": 2.41, "r_error": "", "src_len": 16413, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 8763, "ts": 1791535951.1807745, "chk": "ok", "code_len": 16413}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19047, "tout": 7031, "ts": 1791536018.902929, "chk": null, "code_len": 16413}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791535886.730802}
{"ep": "h2-00852", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0012, "speedup": 1.0012, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 51.0, "calls": 2, "tokens_in": 11415, "tokens_out": 1119, "ctx_chars": 9028, "t_end": 1791536097.978221, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0012, "lo95": 1.001, "mean": 1.0014, "ev_only": 1.0035, "ref_ms": 7.96962, "cand_ms": 7.95987, "n_trials": 150, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 31.78, "compile_s": 1.89, "r_error": "", "src_len": 1029, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 565, "ts": 1791536052.9286027, "chk": "ok", "code_len": 1062}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6046, "tout": 554, "ts": 1791536066.045258, "chk": null, "code_len": 1029}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791536047.5142584}
{"ep": "h2-00817", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0073, "speedup": 1.0073, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 153.5, "calls": 2, "tokens_in": 15137, "tokens_out": 5604, "ctx_chars": 10823, "t_end": 1791536100.076615, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0073, "lo95": 1.0066, "mean": 1.0077, "ev_only": 1.008, "ref_ms": 30.08373, "cand_ms": 29.86518, "n_trials": 48, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0205, "r_wall_s": 83.76, "compile_s": 1.85, "r_error": "", "src_len": 3659, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 3769, "ts": 1791535986.8874364, "chk": "ok", "code_len": 3659}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8782, "tout": 1835, "ts": 1791536015.694879, "chk": null, "code_len": 3659}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791535946.6138139}
{"ep": "h2-00820", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0024, "speedup": 1.0024, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 151.4, "calls": 2, "tokens_in": 11396, "tokens_out": 1094, "ctx_chars": 9028, "t_end": 1791536106.9675782, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0024, "lo95": 1.0022, "mean": 1.0244, "ev_only": 1.0024, "ref_ms": 8.02117, "cand_ms": 8.00174, "n_trials": 175, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0082, "r_wall_s": 103.29, "compile_s": 3.79, "r_error": "", "src_len": 1024, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 546, "ts": 1791535960.9186516, "chk": "ok", "code_len": 1024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6027, "tout": 548, "ts": 1791536003.6022735, "chk": null, "code_len": 1024}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791535955.5926194}
{"ep": "h2-00804", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.399, "speedup": 1.399, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 185.0, "calls": 2, "tokens_in": 6632, "tokens_out": 6816, "ctx_chars": 0, "t_end": 1791536107.1912353, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.399, "lo95": 1.3983, "mean": 1.3983, "ev_only": 1.3993, "ref_ms": 31.3858, "cand_ms": 22.4351, "n_trials": 37, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9997, "r_wall_s": 87.68, "compile_s": 1.95, "r_error": "", "src_len": 3539, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 4928, "ts": 1791535966.4250133, "chk": "ok", "code_len": 3539}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4554, "tout": 1888, "ts": 1791536018.7546427, "chk": null, "code_len": 3539}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791535922.2105117}
{"ep": "h2-00811", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0205, "speedup": 1.0205, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 171.2, "calls": 4, "tokens_in": 34968, "tokens_out": 3687, "ctx_chars": 11823, "t_end": 1791536108.5117986, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0205, "lo95": 1.0195, "mean": 1.0204, "ev_only": 1.0199, "ref_ms": 27.73264, "cand_ms": 27.17574, "n_trials": 53, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0145, "r_wall_s": 74.03, "compile_s": 2.78, "r_error": "", "src_len": 2075, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1487, "ts": 1791535954.6204038, "chk": "ok", "code_len": 2496}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8520, "tout": 1106, "ts": 1791535975.858216, "chk": "ok", "code_len": 2075}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9738, "tout": 27, "ts": 1791536025.5064445, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 9789, "tout": 1067, "ts": 1791536034.4562821, "chk": null, "code_len": 2075}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 6921, "started": 1791535937.298107}
{"ep": "h2-00845", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0106, "speedup": 1.0106, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.2, "calls": 1, "tokens_in": 1599, "tokens_out": 628, "ctx_chars": 0, "t_end": 1791536110.706602, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0106, "lo95": 1.0101, "mean": 1.0105, "ev_only": 1.0115, "ref_ms": 7.98538, "cand_ms": 7.90136, "n_trials": 169, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0044, "r_wall_s": 81.13, "compile_s": 2.2, "r_error": "", "src_len": 1117, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1599, "tout": 628, "ts": 1791536029.2809186, "chk": null, "code_len": 1117}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1599, "started": 1791536018.517508}
{"ep": "h2-00823", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0109, "speedup": 1.0109, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.1, "calls": 2, "tokens_in": 13647, "tokens_out": 1115, "ctx_chars": 10909, "t_end": 1791536113.1926591, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0109, "lo95": 1.0058, "mean": 1.0065, "ev_only": 1.0093, "ref_ms": 8.05607, "cand_ms": 7.96926, "n_trials": 108, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 93.34, "compile_s": 2.39, "r_error": "", "src_len": 938, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 580, "ts": 1791535964.4688647, "chk": "ok", "code_len": 938}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7168, "tout": 535, "ts": 1791536019.5529003, "chk": null, "code_len": 938}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791535959.0576074}
{"ep": "h2-00750", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 7.688, "speedup": 7.688, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 334.6, "calls": 3, "tokens_in": 29570, "tokens_out": 21965, "ctx_chars": 8224, "t_end": 1791536118.5483227, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 7.688, "lo95": 7.5905, "mean": 8.5303, "ev_only": 7.7234, "ref_ms": 32.44043, "cand_ms": 4.21959, "n_trials": 46, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 93.38, "compile_s": 18.52, "r_error": "", "src_len": 8039, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5510, "tout": 13440, "ts": 1791535904.1652021, "chk": "no_custom_kernel", "code_len": 7947}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9927, "tout": 4603, "ts": 1791535983.4002476, "chk": "ok", "code_len": 8039}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 14133, "tout": 3922, "ts": 1791536025.1260004, "chk": null, "code_len": 8039}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5510, "started": 1791535783.9958704}
{"ep": "h2-00805", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0292, "speedup": 1.0292, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 199.4, "calls": 2, "tokens_in": 16123, "tokens_out": 13483, "ctx_chars": 6513, "t_end": 1791536124.890582, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0292, "lo95": 0.7721, "mean": 0.8996, "ev_only": 1.0285, "ref_ms": 8.7354, "cand_ms": 8.4874, "n_trials": 92, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3038, "r_wall_s": 82.04, "compile_s": 2.96, "r_error": "", "src_len": 9513, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 9141, "ts": 1791536004.2643344, "chk": "ok", "code_len": 9513}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10476, "tout": 4342, "ts": 1791536042.7967262, "chk": null, "code_len": 9513}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791535925.5025887}
{"ep": "h2-00851", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0334, "speedup": 1.0334, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 87.1, "calls": 2, "tokens_in": 4968, "tokens_out": 2857, "ctx_chars": 0, "t_end": 1791536126.9650755, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0334, "lo95": 1.0293, "mean": 1.0415, "ev_only": 1.0411, "ref_ms": 8.4218, "cand_ms": 8.14955, "n_trials": 75, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 54.9, "compile_s": 2.02, "r_error": "", "src_len": 1848, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 1874, "ts": 1791536056.0045326, "chk": "ok", "code_len": 1848}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3202, "tout": 983, "ts": 1791536071.9665413, "chk": null, "code_len": 1848}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791536039.852627}
{"ep": "h2-00838", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8441, "speedup": 0.8441, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 123.7, "calls": 2, "tokens_in": 15651, "tokens_out": 2905, "ctx_chars": 11175, "t_end": 1791536127.218215, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.8441, "lo95": 0.8431, "mean": 0.8443, "ev_only": 0.8454, "ref_ms": 14.28339, "cand_ms": 16.92207, "n_trials": 73, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 91.0, "compile_s": 3.8, "r_error": "", "src_len": 2309, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1708, "ts": 1791536018.0214722, "chk": "ok", "code_len": 2309}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8619, "tout": 1197, "ts": 1791536036.0805404, "chk": null, "code_len": 2309}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791536003.5417104}
{"ep": "h2-00784", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.4489, "speedup": 2.4489, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 236.1, "calls": 2, "tokens_in": 29248, "tokens_out": 18802, "ctx_chars": 0, "t_end": 1791536127.7778938, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 2.4489, "lo95": 2.1923, "mean": 2.0018, "ev_only": 2.458, "ref_ms": 26.8012, "cand_ms": 10.94405, "n_trials": 86, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1442, "r_wall_s": 60.57, "compile_s": 2.04, "r_error": "", "src_len": 16539, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 11381, "ts": 1791535984.1098988, "chk": "ok", "code_len": 16539}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18614, "tout": 7421, "ts": 1791536067.1423113, "chk": null, "code_len": 16539}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791535891.726742}
{"ep": "h2-00854", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9954, "speedup": 0.9954, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 76.4, "calls": 2, "tokens_in": 14436, "tokens_out": 3021, "ctx_chars": 10823, "t_end": 1791536131.1143737, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9954, "lo95": 0.9951, "mean": 0.9953, "ev_only": 0.9955, "ref_ms": 30.02948, "cand_ms": 30.16977, "n_trials": 49, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0389, "r_wall_s": 35.85, "compile_s": 1.93, "r_error": "", "src_len": 2872, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1618, "ts": 1791536069.5571826, "chk": "ok", "code_len": 2872}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8083, "tout": 1403, "ts": 1791536094.8470364, "chk": null, "code_len": 2872}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6353, "started": 1791536054.752795}
{"ep": "h2-00814", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.6288, "speedup": 0.6288, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 196.4, "calls": 2, "tokens_in": 15286, "tokens_out": 12531, "ctx_chars": 6513, "t_end": 1791536138.2503088, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.6288, "lo95": 0.6212, "mean": 0.6555, "ev_only": 0.6264, "ref_ms": 7.03346, "cand_ms": 11.18608, "n_trials": 94, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3449, "r_wall_s": 54.61, "compile_s": 2.61, "r_error": "", "src_len": 7361, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 9054, "ts": 1791536021.4722488, "chk": "ok", "code_len": 7436}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9639, "tout": 3477, "ts": 1791536083.5258965, "chk": null, "code_len": 7361}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791535941.8116105}
{"ep": "h2-00864", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0026, "speedup": 1.0026, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 62.7, "calls": 1, "tokens_in": 6367, "tokens_out": 546, "ctx_chars": 10847, "t_end": 1791536141.501417, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0026, "lo95": 1.001, "mean": 1.0025, "ev_only": 1.0053, "ref_ms": 7.98074, "cand_ms": 7.96026, "n_trials": 102, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0047, "r_wall_s": 57.28, "compile_s": 1.96, "r_error": "", "src_len": 1029, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 546, "ts": 1791536084.1426182, "chk": null, "code_len": 1029}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791536078.782612}
{"ep": "h2-00786", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7003, "speedup": 0.7003, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 250.3, "calls": 2, "tokens_in": 13774, "tokens_out": 13841, "ctx_chars": 6513, "t_end": 1791536146.0544655, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7003, "lo95": 0.7, "mean": 0.7005, "ev_only": 0.7, "ref_ms": 15.73016, "cand_ms": 22.46208, "n_trials": 65, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.2669, "r_wall_s": 30.79, "compile_s": 2.07, "r_error": "", "src_len": 7117, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 10360, "ts": 1791536031.218962, "chk": "ok", "code_len": 7117}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8910, "tout": 3481, "ts": 1791536114.9171913, "chk": null, "code_len": 7117}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791535895.7388604}
{"ep": "h2-00788", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.3179, "speedup": 0.3179, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 248.4, "calls": 5, "tokens_in": 56572, "tokens_out": 11744, "ctx_chars": 6513, "t_end": 1791536146.0579655, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.3179, "lo95": 0.2535, "mean": 0.3266, "ev_only": 0.3169, "ref_ms": 10.09197, "cand_ms": 31.74295, "n_trials": 41, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0526, "r_wall_s": 36.62, "compile_s": 1.92, "r_error": "", "src_len": 6152, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2958, "ts": 1791535923.6610093, "chk": "incorrect", "code_len": 6698}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8735, "tout": 3062, "ts": 1791535957.8666573, "chk": "incorrect", "code_len": 6573}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 11929, "tout": 3040, "ts": 1791536032.0899794, "chk": "ok", "code_len": 6152}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 15081, "tout": 54, "ts": 1791536088.210135, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 15182, "tout": 2630, "ts": 1791536109.409341, "chk": null, "code_len": 6152}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791535897.6408029}
{"ep": "h2-00869", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0011, "speedup": 1.0011, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 53.6, "calls": 3, "tokens_in": 17512, "tokens_out": 1124, "ctx_chars": 9028, "t_end": 1791536146.7787971, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0011, "lo95": 1.0003, "mean": 1.0049, "ev_only": 1.0044, "ref_ms": 8.01165, "cand_ms": 8.00307, "n_trials": 191, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0067, "r_wall_s": 33.62, "compile_s": 2.08, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 572, "ts": 1791536098.8561752, "chk": "ok", "code_len": 1063}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6053, "tout": 13, "ts": 1791536108.339968, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6090, "tout": 539, "ts": 1791536113.0895686, "chk": null, "code_len": 999}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5369, "started": 1791536093.148417}
{"ep": "h2-00790", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.5373, "speedup": 1.5373, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 244.8, "calls": 3, "tokens_in": 15931, "tokens_out": 12766, "ctx_chars": 0, "t_end": 1791536146.8780797, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.5373, "lo95": 1.537, "mean": 1.535, "ev_only": 1.5362, "ref_ms": 15.47345, "cand_ms": 10.06525, "n_trials": 72, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.2775, "r_wall_s": 108.46, "compile_s": 15.12, "r_error": "", "src_len": 6465, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1904, "tout": 6578, "ts": 1791535960.3000445, "chk": "forbidden", "code_len": 6681}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5434, "tout": 3177, "ts": 1791535982.1430085, "chk": "ok", "code_len": 6465}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8593, "tout": 3011, "ts": 1791536037.7347524, "chk": null, "code_len": 6465}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", null], "first_tin": 1904, "started": 1791535902.0325394}
{"ep": "h2-00848", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0134, "speedup": 1.0134, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 123.9, "calls": 3, "tokens_in": 9157, "tokens_out": 3010, "ctx_chars": 0, "t_end": 1791536149.0950367, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0134, "lo95": 1.0128, "mean": 1.0214, "ev_only": 1.0132, "ref_ms": 64.28908, "cand_ms": 63.44054, "n_trials": 21, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0029, "r_wall_s": 36.89, "compile_s": 1.98, "r_error": "", "src_len": 2399, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1744, "tout": 1810, "ts": 1791536065.6694095, "chk": "ok", "code_len": 2658}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3666, "tout": 57, "ts": 1791536074.6957803, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 3747, "tout": 1143, "ts": 1791536112.1822867, "chk": null, "code_len": 2399}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1744, "started": 1791536025.1721623}
{"ep": "h2-00826", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.5464, "speedup": 1.5464, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 188.6, "calls": 2, "tokens_in": 5192, "tokens_out": 4217, "ctx_chars": 0, "t_end": 1791536150.2544696, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.5464, "lo95": 1.5457, "mean": 1.546, "ev_only": 1.5488, "ref_ms": 7.97099, "cand_ms": 5.1544, "n_trials": 183, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 111.45, "compile_s": 3.65, "r_error": "", "src_len": 2705, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 2669, "ts": 1791535984.3830986, "chk": "ok", "code_len": 2705}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3591, "tout": 1548, "ts": 1791536038.7175322, "chk": null, "code_len": 2705}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791535961.6718426}
{"ep": "h2-00720", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.2575, "speedup": 0.2575, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 444.5, "calls": 3, "tokens_in": 16284, "tokens_out": 14756, "ctx_chars": 0, "t_end": 1791536150.553573, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.2575, "lo95": 0.2572, "mean": 0.2581, "ev_only": 0.2577, "ref_ms": 8.5039, "cand_ms": 33.02793, "n_trials": 43, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.9997, "r_wall_s": 118.93, "compile_s": 54.34, "r_error": "", "src_len": 6902, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 7781, "ts": 1791535775.7177145, "chk": "no_custom_kernel", "code_len": 6460}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5492, "tout": 3697, "ts": 1791535913.842973, "chk": "ok", "code_len": 6902}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8972, "tout": 3278, "ts": 1791536031.5607898, "chk": null, "code_len": 6902}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1820, "started": 1791535706.0245397}
{"ep": "h2-00819", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0045, "speedup": 1.0045, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 199.7, "calls": 3, "tokens_in": 24058, "tokens_out": 2760, "ctx_chars": 11823, "t_end": 1791536153.6768746, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0045, "lo95": 1.0036, "mean": 1.005, "ev_only": 1.0046, "ref_ms": 27.46621, "cand_ms": 27.34387, "n_trials": 51, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0282, "r_wall_s": 109.8, "compile_s": 2.78, "r_error": "", "src_len": 2516, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1517, "ts": 1791535968.4390762, "chk": "ok", "code_len": 2634}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8550, "tout": 13, "ts": 1791536033.833547, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8587, "tout": 1230, "ts": 1791536043.8357725, "chk": null, "code_len": 2516}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6921, "started": 1791535953.9493964}
{"ep": "h2-00785", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.5181, "speedup": 0.5181, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 260.7, "calls": 3, "tokens_in": 13248, "tokens_out": 6405, "ctx_chars": 0, "t_end": 1791536154.7676828, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.5181, "lo95": 0.5179, "mean": 0.5181, "ev_only": 0.518, "ref_ms": 31.4799, "cand_ms": 60.75952, "n_trials": 22, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 81.41, "compile_s": 2.19, "r_error": "", "src_len": 3986, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2076, "tout": 2287, "ts": 1791535915.795486, "chk": "incorrect", "code_len": 4119}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4496, "tout": 2071, "ts": 1791535972.5754619, "chk": "ok", "code_len": 3902}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6676, "tout": 2047, "ts": 1791536072.564402, "chk": null, "code_len": 3986}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", null], "first_tin": 2076, "started": 1791535894.0358572}
{"ep": "h2-00856", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0563, "speedup": 1.0563, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 101.6, "calls": 2, "tokens_in": 12974, "tokens_out": 2280, "ctx_chars": 9274, "t_end": 1791536159.5095217, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0563, "lo95": 1.0464, "mean": 1.0397, "ev_only": 1.0524, "ref_ms": 8.45526, "cand_ms": 8.00455, "n_trials": 155, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 73.67, "compile_s": 1.93, "r_error": "", "src_len": 1802, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1278, "ts": 1791536068.2203503, "chk": "ok", "code_len": 1802}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7182, "tout": 1002, "ts": 1791536085.767906, "chk": null, "code_len": 1802}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791536057.8630702}
{"ep": "h2-00862", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8483, "speedup": 0.8483, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 84.8, "calls": 1, "tokens_in": 7030, "tokens_out": 1207, "ctx_chars": 11175, "t_end": 1791536161.01058, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.8483, "lo95": 0.8469, "mean": 0.8719, "ev_only": 0.8471, "ref_ms": 14.42615, "cand_ms": 17.00556, "n_trials": 62, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 73.4, "compile_s": 1.99, "r_error": "", "src_len": 2585, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1207, "ts": 1791536087.4320266, "chk": null, "code_len": 2585}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791536076.245118}
{"ep": "h2-00867", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0278, "speedup": 1.0278, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.9, "calls": 1, "tokens_in": 5832, "tokens_out": 526, "ctx_chars": 9656, "t_end": 1791536169.1410143, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0278, "lo95": 1.0265, "mean": 1.0279, "ev_only": 1.0274, "ref_ms": 24.86565, "cand_ms": 24.19256, "n_trials": 45, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0223, "r_wall_s": 71.83, "compile_s": 2.17, "r_error": "", "src_len": 1002, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 526, "ts": 1791536096.5818968, "chk": null, "code_len": 1002}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791536091.235183}
{"ep": "h2-00858", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0667, "speedup": 1.0667, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 106.1, "calls": 2, "tokens_in": 10493, "tokens_out": 2282, "ctx_chars": 6513, "t_end": 1791536169.3790946, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0667, "lo95": 1.0664, "mean": 1.066, "ev_only": 1.0657, "ref_ms": 15.46933, "cand_ms": 14.50237, "n_trials": 96, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1326, "r_wall_s": 69.97, "compile_s": 2.17, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1171, "ts": 1791536073.332491, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5888, "tout": 1111, "ts": 1791536098.2422893, "chk": null, "code_len": 2422}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791536063.2521486}
{"ep": "h2-00855", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0152, "speedup": 1.0152, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 126.5, "calls": 2, "tokens_in": 16181, "tokens_out": 2585, "ctx_chars": 12117, "t_end": 1791536181.338926, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0152, "lo95": 1.0137, "mean": 1.0187, "ev_only": 1.0155, "ref_ms": 42.86788, "cand_ms": 42.22706, "n_trials": 38, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 89.34, "compile_s": 1.98, "r_error": "", "src_len": 2064, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1542, "ts": 1791536067.640227, "chk": "ok", "code_len": 2064}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8782, "tout": 1043, "ts": 1791536091.3328156, "chk": null, "code_len": 2064}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791536054.846851}
{"ep": "h2-00830", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9898, "speedup": 0.9898, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 212.3, "calls": 3, "tokens_in": 11764, "tokens_out": 3019, "ctx_chars": 0, "t_end": 1791536187.3830721, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9898, "lo95": 0.9893, "mean": 0.9896, "ev_only": 0.9896, "ref_ms": 42.2858, "cand_ms": 42.7225, "n_trials": 34, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.0117, "r_wall_s": 94.63, "compile_s": 11.58, "r_error": "", "src_len": 1632, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2034, "tout": 1146, "ts": 1791535987.9064002, "chk": "compile_error", "code_len": 1573}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4432, "tout": 754, "ts": 1791536057.1505032, "chk": "ok", "code_len": 1632}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5298, "tout": 1119, "ts": 1791536092.002796, "chk": null, "code_len": 1632}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", null], "first_tin": 2034, "started": 1791535975.1236246}
{"ep": "h2-00875", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8201, "speedup": 0.8201, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.1, "calls": 2, "tokens_in": 12029, "tokens_out": 1448, "ctx_chars": 8224, "t_end": 1791536188.2691314, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8201, "lo95": 0.8147, "mean": 0.8828, "ev_only": 0.8199, "ref_ms": 32.8646, "cand_ms": 40.0746, "n_trials": 37, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 14.79, "compile_s": 2.13, "r_error": "", "src_len": 999, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 904, "ts": 1791536119.850837, "chk": "ok", "code_len": 999}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6521, "tout": 544, "ts": 1791536173.4379826, "chk": null, "code_len": 999}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5508, "started": 1791536107.72208}
{"ep": "h2-00843", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9776, "speedup": 0.9776, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 175.5, "calls": 2, "tokens_in": 15210, "tokens_out": 1836, "ctx_chars": 11539, "t_end": 1791536188.756122, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9776, "lo95": 0.9771, "mean": 0.9928, "ev_only": 0.9764, "ref_ms": 31.488, "cand_ms": 32.21062, "n_trials": 41, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0185, "r_wall_s": 102.76, "compile_s": 2.53, "r_error": "", "src_len": 1761, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1037, "ts": 1791536028.0777853, "chk": "ok", "code_len": 1761}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8178, "tout": 799, "ts": 1791536085.075789, "chk": null, "code_len": 1761}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791536013.2748005}
{"ep": "h2-00846", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.03, "speedup": 1.03, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 170.3, "calls": 2, "tokens_in": 8884, "tokens_out": 7891, "ctx_chars": 0, "t_end": 1791536189.8129241, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.03, "lo95": 1.026, "mean": 1.1023, "ev_only": 1.0318, "ref_ms": 3.28088, "cand_ms": 3.18542, "n_trials": 208, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0887, "r_wall_s": 90.22, "compile_s": 2.84, "r_error": "", "src_len": 6269, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2745, "tout": 4770, "ts": 1791536067.5420153, "chk": "ok", "code_len": 6269}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6139, "tout": 3121, "ts": 1791536099.4855804, "chk": null, "code_len": 6269}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2745, "started": 1791536019.5368507}
{"ep": "h2-00844", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0239, "speedup": 1.0239, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 180.7, "calls": 4, "tokens_in": 19353, "tokens_out": 4784, "ctx_chars": 0, "t_end": 1791536197.131787, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0239, "lo95": 1.0237, "mean": 1.0238, "ev_only": 1.0238, "ref_ms": 8.28409, "cand_ms": 8.09104, "n_trials": 146, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0249, "r_wall_s": 24.48, "compile_s": 11.12, "r_error": "", "src_len": 3573, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1764, "tout": 1741, "ts": 1791536054.897034, "chk": "compile_error", "code_len": 3779}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4759, "tout": 1505, "ts": 1791536121.2968693, "chk": "ok", "code_len": 3573}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6376, "tout": 54, "ts": 1791536159.0164988, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6454, "tout": 1484, "ts": 1791536172.5487876, "chk": null, "code_len": 3573}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 1764, "started": 1791536016.3954825}
{"ep": "h2-00833", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0062, "speedup": 1.0062, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 209.2, "calls": 2, "tokens_in": 5860, "tokens_out": 5947, "ctx_chars": 0, "t_end": 1791536199.7063959, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0062, "lo95": 1.0057, "mean": 1.0062, "ev_only": 1.0065, "ref_ms": 30.07451, "cand_ms": 29.88924, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0163, "r_wall_s": 97.34, "compile_s": 2.53, "r_error": "", "src_len": 3919, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 3978, "ts": 1791536025.0281801, "chk": "ok", "code_len": 3919}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4262, "tout": 1969, "ts": 1791536101.8780377, "chk": null, "code_len": 3919}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791535990.5529826}
{"ep": "h2-00860", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8495, "speedup": 0.8495, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 130.1, "calls": 2, "tokens_in": 15732, "tokens_out": 3041, "ctx_chars": 11175, "t_end": 1791536200.5249517, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8495, "lo95": 0.8478, "mean": 0.849, "ev_only": 0.8485, "ref_ms": 14.35342, "cand_ms": 16.89548, "n_trials": 65, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 26.01, "compile_s": 1.9, "r_error": "", "src_len": 2458, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1824, "ts": 1791536102.1896179, "chk": "ok", "code_len": 2458}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8700, "tout": 1217, "ts": 1791536174.405706, "chk": null, "code_len": 2458}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7032, "started": 1791536070.457793}
{"ep": "h2-00831", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.6182, "speedup": 2.6182, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 222.5, "calls": 3, "tokens_in": 15148, "tokens_out": 12275, "ctx_chars": 0, "t_end": 1791536201.4452565, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.6182, "lo95": 2.6159, "mean": 2.6173, "ev_only": 2.6233, "ref_ms": 14.33576, "cand_ms": 5.47546, "n_trials": 98, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 24.95, "compile_s": 11.55, "r_error": "", "src_len": 5218, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6992, "ts": 1791536064.7362797, "chk": "no_custom_kernel", "code_len": 5217}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5149, "tout": 2814, "ts": 1791536100.6301477, "chk": "ok", "code_len": 5218}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7816, "tout": 2469, "ts": 1791536176.1696665, "chk": null, "code_len": 5218}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2183, "started": 1791535978.8971038}
{"ep": "h2-00800", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.1705, "speedup": 1.1705, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 289.7, "calls": 2, "tokens_in": 7205, "tokens_out": 8534, "ctx_chars": 0, "t_end": 1791536204.0229182, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.1705, "lo95": 1.1702, "mean": 1.1704, "ev_only": 1.1711, "ref_ms": 31.62797, "cand_ms": 27.02201, "n_trials": 49, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9995, "r_wall_s": 194.79, "compile_s": 2.8, "r_error": "", "src_len": 4702, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 5814, "ts": 1791535965.1082525, "chk": "ok", "code_len": 4702}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5127, "tout": 2720, "ts": 1791536007.9446733, "chk": null, "code_len": 4702}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791535914.3045268}
{"ep": "h2-00837", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8702, "speedup": 0.8702, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 212.0, "calls": 2, "tokens_in": 4765, "tokens_out": 2177, "ctx_chars": 0, "t_end": 1791536211.581051, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8702, "lo95": 0.8691, "mean": 0.8707, "ev_only": 0.8703, "ref_ms": 27.51814, "cand_ms": 31.62294, "n_trials": 46, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1579, "r_wall_s": 69.65, "compile_s": 2.52, "r_error": "", "src_len": 1896, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1728, "tout": 1197, "ts": 1791536026.8466036, "chk": "ok", "code_len": 1971}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3037, "tout": 980, "ts": 1791536141.910583, "chk": null, "code_len": 1896}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1728, "started": 1791535999.6114733}
{"ep": "h2-00807", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0033, "speedup": 1.0033, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 280.8, "calls": 3, "tokens_in": 24687, "tokens_out": 2154, "ctx_chars": 12117, "t_end": 1791536215.824195, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0033, "lo95": 1.0022, "mean": 1.0038, "ev_only": 1.0036, "ref_ms": 42.48384, "cand_ms": 42.34563, "n_trials": 32, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0, "r_wall_s": 185.53, "compile_s": 2.08, "r_error": "", "src_len": 2244, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1097, "ts": 1791535945.1224458, "chk": "ok", "code_len": 2244}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8606, "tout": 54, "ts": 1791536015.9820898, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8684, "tout": 1003, "ts": 1791536029.2331715, "chk": null, "code_len": 2244}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791535934.976022}
{"ep": "h2-00789", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0208, "speedup": 1.0208, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 320.7, "calls": 3, "tokens_in": 16344, "tokens_out": 2826, "ctx_chars": 5026, "t_end": 1791536219.1142695, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0208, "lo95": 1.0184, "mean": 1.0311, "ev_only": 1.0207, "ref_ms": 8.37323, "cand_ms": 8.20227, "n_trials": 156, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0093, "r_wall_s": 175.51, "compile_s": 4.06, "r_error": "", "src_len": 1452, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 2025, "ts": 1791535916.4833827, "chk": "ok", "code_len": 1452}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6130, "tout": 67, "ts": 1791536031.6918678, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6221, "tout": 734, "ts": 1791536043.5125725, "chk": null, "code_len": 1452}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 3993, "started": 1791535898.3780417}
{"ep": "h2-00801", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0214, "speedup": 1.0214, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 300.6, "calls": 3, "tokens_in": 12338, "tokens_out": 8435, "ctx_chars": 0, "t_end": 1791536220.082642, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0214, "lo95": 1.0173, "mean": 1.018, "ev_only": 1.0206, "ref_ms": 30.71588, "cand_ms": 30.07201, "n_trials": 31, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0195, "r_wall_s": 203.12, "compile_s": 2.49, "r_error": "", "src_len": 4112, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 4067, "ts": 1791535960.3016932, "chk": "forbidden", "code_len": 3925}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4230, "tout": 2276, "ts": 1791535976.1381555, "chk": "ok", "code_len": 4112}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6483, "tout": 2092, "ts": 1791536016.5265677, "chk": null, "code_len": 4112}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", null], "first_tin": 1625, "started": 1791535920.1061666}
{"ep": "h2-00821", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.028, "speedup": 1.028, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 269.5, "calls": 2, "tokens_in": 12722, "tokens_out": 1977, "ctx_chars": 9656, "t_end": 1791536225.2844958, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.028, "lo95": 1.0248, "mean": 1.023, "ev_only": 1.0279, "ref_ms": 25.12875, "cand_ms": 24.44479, "n_trials": 24, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.014, "r_wall_s": 199.36, "compile_s": 2.02, "r_error": "", "src_len": 1106, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 1339, "ts": 1791535967.9747941, "chk": "ok", "code_len": 1106}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6888, "tout": 638, "ts": 1791536025.2280831, "chk": null, "code_len": 1106}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791535955.7731204}
{"ep": "h2-00897", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0051, "speedup": 1.0051, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 75.6, "calls": 2, "tokens_in": 14091, "tokens_out": 1481, "ctx_chars": 10909, "t_end": 1791536236.6504548, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0051, "lo95": 1.0047, "mean": 1.0136, "ev_only": 1.0055, "ref_ms": 7.97791, "cand_ms": 7.93707, "n_trials": 163, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 16.87, "compile_s": 1.86, "r_error": "", "src_len": 898, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6477, "tout": 1028, "ts": 1791536173.5420582, "chk": "ok", "code_len": 898}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7614, "tout": 453, "ts": 1791536219.6842759, "chk": null, "code_len": 898}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6477, "started": 1791536161.0108452}
{"ep": "h2-00868", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0651, "speedup": 1.0651, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 155.9, "calls": 2, "tokens_in": 10519, "tokens_out": 2332, "ctx_chars": 6513, "t_end": 1791536248.0692613, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0651, "lo95": 1.0645, "mean": 1.0667, "ev_only": 1.0653, "ref_ms": 15.4741, "cand_ms": 14.52885, "n_trials": 109, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1319, "r_wall_s": 64.61, "compile_s": 1.94, "r_error": "", "src_len": 2454, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1197, "ts": 1791536102.7704463, "chk": "ok", "code_len": 2454}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5914, "tout": 1135, "ts": 1791536182.7854416, "chk": null, "code_len": 2454}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791536092.1509774}
{"ep": "h2-00895", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.712, "speedup": 0.712, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 93.4, "calls": 2, "tokens_in": 14872, "tokens_out": 9052, "ctx_chars": 6513, "t_end": 1791536248.1451926, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.712, "lo95": 0.7094, "mean": 0.7129, "ev_only": 0.7114, "ref_ms": 3.23376, "cand_ms": 4.54161, "n_trials": 182, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0981, "r_wall_s": 14.86, "compile_s": 2.02, "r_error": "", "src_len": 7354, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 5601, "ts": 1791536200.1522067, "chk": "ok", "code_len": 7354}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9424, "tout": 3451, "ts": 1791536233.2381506, "chk": null, "code_len": 7354}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791536154.7680774}
{"ep": "h2-00815", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.5011, "speedup": 0.5011, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 308.7, "calls": 4, "tokens_in": 19995, "tokens_out": 6363, "ctx_chars": 0, "t_end": 1791536250.686511, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.5011, "lo95": 0.5008, "mean": 0.5014, "ev_only": 0.5009, "ref_ms": 31.39897, "cand_ms": 62.65765, "n_trials": 22, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 126.93, "compile_s": 2.82, "r_error": "", "src_len": 3864, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2076, "tout": 2229, "ts": 1791535963.9461453, "chk": "incorrect", "code_len": 4029}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4438, "tout": 2169, "ts": 1791536043.4375947, "chk": "ok", "code_len": 3864}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6716, "tout": 25, "ts": 1791536087.4050658, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 6765, "tout": 1940, "ts": 1791536122.4748697, "chk": null, "code_len": 3864}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "no_code", null], "first_tin": 2076, "started": 1791535941.99061}
{"ep": "h2-00886", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.3289, "speedup": 0.3289, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 116.2, "calls": 3, "tokens_in": 22629, "tokens_out": 5955, "ctx_chars": 6513, "t_end": 1791536257.6515367, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.3289, "lo95": 0.3265, "mean": 0.321, "ev_only": 0.3273, "ref_ms": 3.32228, "cand_ms": 10.10122, "n_trials": 86, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0261, "r_wall_s": 51.32, "compile_s": 1.95, "r_error": "", "src_len": 6000, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 3008, "ts": 1791536170.1108925, "chk": "ok", "code_len": 6024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8566, "tout": 27, "ts": 1791536181.2168996, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8617, "tout": 2920, "ts": 1791536206.3039093, "chk": null, "code_len": 6000}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791536141.5016613}
{"ep": "h2-00859", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.323, "speedup": 0.323, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 193.4, "calls": 4, "tokens_in": 35337, "tokens_out": 7541, "ctx_chars": 6513, "t_end": 1791536258.3316047, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.323, "lo95": 0.3201, "mean": 0.3173, "ev_only": 0.3212, "ref_ms": 3.30333, "cand_ms": 10.22678, "n_trials": 152, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0263, "r_wall_s": 38.13, "compile_s": 2.6, "r_error": "", "src_len": 5231, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2505, "ts": 1791536085.6510055, "chk": "forbidden", "code_len": 5241}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8189, "tout": 2524, "ts": 1791536183.4068015, "chk": "ok", "code_len": 5231}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10825, "tout": 28, "ts": 1791536197.9449914, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 10877, "tout": 2484, "ts": 1791536219.8915246, "chk": null, "code_len": 5231}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", "no_code", null], "first_tin": 5446, "started": 1791536064.919702}
{"ep": "h2-00885", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.3464, "speedup": 2.3464, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 123.3, "calls": 3, "tokens_in": 23560, "tokens_out": 5180, "ctx_chars": 8529, "t_end": 1791536261.5347896, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.3464, "lo95": 2.3436, "mean": 2.343, "ev_only": 2.348, "ref_ms": 8.53105, "cand_ms": 3.63577, "n_trials": 180, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.5389, "r_wall_s": 40.68, "compile_s": 2.39, "r_error": "", "src_len": 3388, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 3399, "ts": 1791536177.5673165, "chk": "ok", "code_len": 3388}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9006, "tout": 29, "ts": 1791536205.6978369, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9059, "tout": 1752, "ts": 1791536220.7952783, "chk": null, "code_len": 3388}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5495, "started": 1791536138.250585}
{"ep": "h2-00894", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.316, "speedup": 0.316, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.0, "calls": 3, "tokens_in": 22429, "tokens_out": 5801, "ctx_chars": 6513, "t_end": 1791536264.6689131, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.316, "lo95": 0.2988, "mean": 0.3257, "ev_only": 0.3148, "ref_ms": 3.35392, "cand_ms": 10.61274, "n_trials": 127, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0256, "r_wall_s": 42.87, "compile_s": 1.93, "r_error": "", "src_len": 5895, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2915, "ts": 1791536177.3072631, "chk": "ok", "code_len": 5895}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8473, "tout": 13, "ts": 1791536197.9628816, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8510, "tout": 2873, "ts": 1791536221.5513263, "chk": null, "code_len": 5895}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791536153.677121}
{"ep": "h2-00879", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.067, "speedup": 1.067, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 151.6, "calls": 1, "tokens_in": 4605, "tokens_out": 1150, "ctx_chars": 6513, "t_end": 1791536270.142812, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.067, "lo95": 1.0658, "mean": 1.0633, "ev_only": 1.0657, "ref_ms": 15.50671, "cand_ms": 14.53341, "n_trials": 106, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1489, "r_wall_s": 140.95, "compile_s": 4.88, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 4605, "tout": 1150, "ts": 1791536128.4133594, "chk": null, "code_len": 2422}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 4605, "started": 1791536118.5485568}
{"ep": "h2-00907", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0198, "speedup": 1.0198, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.5, "calls": 2, "tokens_in": 15640, "tokens_out": 3640, "ctx_chars": 11823, "t_end": 1791536278.0028481, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0198, "lo95": 1.0185, "mean": 1.0575, "ev_only": 1.0189, "ref_ms": 27.68462, "cand_ms": 27.14718, "n_trials": 41, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0119, "r_wall_s": 31.92, "compile_s": 2.21, "r_error": "", "src_len": 2261, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 2372, "ts": 1791536223.7329702, "chk": "ok", "code_len": 2261}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8717, "tout": 1268, "ts": 1791536246.0471869, "chk": null, "code_len": 2261}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791536200.525185}
{"ep": "h2-00825", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4324, "speedup": 1.4324, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 323.1, "calls": 2, "tokens_in": 16620, "tokens_out": 14176, "ctx_chars": 6513, "t_end": 1791536283.864276, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.4324, "lo95": 1.4205, "mean": 1.4285, "ev_only": 1.4281, "ref_ms": 3.32368, "cand_ms": 2.32031, "n_trials": 177, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1399, "r_wall_s": 52.99, "compile_s": 43.39, "r_error": "", "src_len": 11172, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 8976, "ts": 1791536034.3656015, "chk": "ok", "code_len": 11172}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 11172, "tout": 5200, "ts": 1791536230.6254907, "chk": null, "code_len": 11172}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791535960.7221568}
{"ep": "h2-00892", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0302, "speedup": 1.0302, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 134.0, "calls": 2, "tokens_in": 9235, "tokens_out": 2587, "ctx_chars": 5026, "t_end": 1791536284.3006232, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0302, "lo95": 1.0235, "mean": 1.024, "ev_only": 1.0265, "ref_ms": 8.25049, "cand_ms": 8.00866, "n_trials": 132, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0066, "r_wall_s": 39.89, "compile_s": 1.99, "r_error": "", "src_len": 1331, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1797, "ts": 1791536166.4730666, "chk": "ok", "code_len": 1331}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5240, "tout": 790, "ts": 1791536244.247839, "chk": null, "code_len": 1331}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791536150.2547216}
{"ep": "h2-00905", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0272, "speedup": 1.0272, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 88.3, "calls": 2, "tokens_in": 12339, "tokens_out": 1087, "ctx_chars": 9656, "t_end": 1791536285.3931308, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0272, "lo95": 1.0261, "mean": 1.0273, "ev_only": 1.0275, "ref_ms": 24.82032, "cand_ms": 24.16404, "n_trials": 52, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0287, "r_wall_s": 63.02, "compile_s": 2.54, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5832, "tout": 563, "ts": 1791536202.4624326, "chk": "ok", "code_len": 996}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6507, "tout": 524, "ts": 1791536221.5953555, "chk": null, "code_len": 996}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5832, "started": 1791536197.132022}
{"ep": "h2-00899", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0544, "speedup": 1.0544, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 117.1, "calls": 2, "tokens_in": 4618, "tokens_out": 2790, "ctx_chars": 0, "t_end": 1791536286.4961584, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0544, "lo95": 1.0536, "mean": 1.056, "ev_only": 1.0547, "ref_ms": 24.84139, "cand_ms": 23.55885, "n_trials": 44, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.025, "r_wall_s": 79.32, "compile_s": 1.97, "r_error": "", "src_len": 1423, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1978, "ts": 1791536186.386982, "chk": "ok", "code_len": 1423}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2950, "tout": 812, "ts": 1791536206.380995, "chk": null, "code_len": 1423}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791536169.3793273}
{"ep": "h2-00906", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0264, "speedup": 1.0264, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 90.3, "calls": 2, "tokens_in": 12320, "tokens_out": 1068, "ctx_chars": 9656, "t_end": 1791536289.9987795, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0264, "lo95": 1.0254, "mean": 1.0261, "ev_only": 1.0267, "ref_ms": 24.89069, "cand_ms": 24.25063, "n_trials": 60, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0286, "r_wall_s": 65.15, "compile_s": 2.58, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5832, "tout": 544, "ts": 1791536205.1219573, "chk": "ok", "code_len": 996}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6488, "tout": 524, "ts": 1791536223.9488354, "chk": null, "code_len": 996}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5832, "started": 1791536199.706641}
{"ep": "h2-00883", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0107, "speedup": 1.0107, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 167.9, "calls": 2, "tokens_in": 6805, "tokens_out": 8326, "ctx_chars": 0, "t_end": 1791536295.6722782, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0107, "lo95": 1.0098, "mean": 1.0147, "ev_only": 1.0104, "ref_ms": 30.10057, "cand_ms": 29.78218, "n_trials": 37, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0175, "r_wall_s": 77.03, "compile_s": 17.24, "r_error": "", "src_len": 6094, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 5386, "ts": 1791536175.2617192, "chk": "ok", "code_len": 6094}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5207, "tout": 2940, "ts": 1791536218.2162623, "chk": null, "code_len": 6094}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791536127.7781224}
{"ep": "h2-00891", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0115, "speedup": 1.0115, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 148.1, "calls": 2, "tokens_in": 16066, "tokens_out": 2245, "ctx_chars": 12117, "t_end": 1791536297.2031114, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0115, "lo95": 1.0104, "mean": 1.0111, "ev_only": 1.0109, "ref_ms": 42.60969, "cand_ms": 42.12332, "n_trials": 35, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 92.35, "compile_s": 1.82, "r_error": "", "src_len": 2326, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1160, "ts": 1791536159.533046, "chk": "ok", "code_len": 2326}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8669, "tout": 1085, "ts": 1791536204.1629925, "chk": null, "code_len": 2326}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7397, "started": 1791536149.0952682}
{"ep": "h2-00816", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 6.9559, "speedup": 6.9559, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 367.0, "calls": 3, "tokens_in": 20443, "tokens_out": 24451, "ctx_chars": 0, "t_end": 1791536312.4548416, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 6.9559, "lo95": 6.936, "mean": 7.2289, "ev_only": 6.9814, "ref_ms": 33.34318, "cand_ms": 4.79354, "n_trials": 46, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9668, "r_wall_s": 55.05, "compile_s": 15.47, "r_error": "", "src_len": 9122, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 15289, "ts": 1791536081.2587526, "chk": "no_custom_kernel", "code_len": 9179}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6964, "tout": 4863, "ts": 1791536146.786199, "chk": "ok", "code_len": 9122}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11513, "tout": 4299, "ts": 1791536257.374815, "chk": null, "code_len": 9122}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791535945.4375374}
{"ep": "h2-00908", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9756, "speedup": 0.9756, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.2, "calls": 3, "tokens_in": 23601, "tokens_out": 1970, "ctx_chars": 11539, "t_end": 1791536312.6375778, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9756, "lo95": 0.9753, "mean": 0.9755, "ev_only": 0.9759, "ref_ms": 31.37824, "cand_ms": 32.16385, "n_trials": 39, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0186, "r_wall_s": 73.68, "compile_s": 1.83, "r_error": "", "src_len": 1798, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1125, "ts": 1791536213.4015718, "chk": "ok", "code_len": 1798}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8266, "tout": 13, "ts": 1791536231.034719, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8303, "tout": 832, "ts": 1791536238.0897481, "chk": null, "code_len": 1798}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7032, "started": 1791536201.4455142}
{"ep": "h2-00835", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.2381, "speedup": 1.2381, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 317.4, "calls": 2, "tokens_in": 12577, "tokens_out": 22296, "ctx_chars": 0, "t_end": 1791536314.1893456, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.2381, "lo95": 1.2326, "mean": 1.1511, "ev_only": 1.2353, "ref_ms": 6.97741, "cand_ms": 5.63539, "n_trials": 197, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.994, "r_wall_s": 59.61, "compile_s": 2.03, "r_error": "", "src_len": 12534, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 16125, "ts": 1791536137.4101558, "chk": "ok", "code_len": 12605}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9633, "tout": 6171, "ts": 1791536254.4247792, "chk": null, "code_len": 12534}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791535996.757053}
{"ep": "h2-00925", "task": "L1_012", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7527, "speedup": 0.7527, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 32.7, "calls": 1, "tokens_in": 6622, "tokens_out": 521, "ctx_chars": 10850, "t_end": 1791536316.5969787, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7527, "lo95": 0.7484, "mean": 0.7472, "ev_only": 0.7599, "ref_ms": 0.15289, "cand_ms": 0.20313, "n_trials": 179, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 16.92, "compile_s": 1.97, "r_error": "", "src_len": 856, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6622, "tout": 521, "ts": 1791536299.3214712, "chk": null, "code_len": 856}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6622, "started": 1791536283.8645396}
{"ep": "h2-00890", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.8917, "speedup": 2.8917, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 171.6, "calls": 3, "tokens_in": 12529, "tokens_out": 7994, "ctx_chars": 0, "t_end": 1791536318.4985116, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.8917, "lo95": 2.7844, "mean": 2.8272, "ev_only": 2.9404, "ref_ms": 1.8664, "cand_ms": 0.64544, "n_trials": 227, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 78.73, "compile_s": 18.21, "r_error": "", "src_len": 3774, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 3864, "ts": 1791536179.9877295, "chk": "no_custom_kernel", "code_len": 3795}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4316, "tout": 2158, "ts": 1791536210.8469381, "chk": "ok", "code_len": 3774}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6454, "tout": 1972, "ts": 1791536239.7410698, "chk": null, "code_len": 3774}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1759, "started": 1791536146.8782814}
{"ep": "h2-00922", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0123, "speedup": 1.0123, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 54.4, "calls": 2, "tokens_in": 11902, "tokens_out": 1619, "ctx_chars": 9028, "t_end": 1791536319.031445, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0123, "lo95": 1.0086, "mean": 1.0079, "ev_only": 1.01, "ref_ms": 8.13077, "cand_ms": 8.03209, "n_trials": 177, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0076, "r_wall_s": 20.41, "compile_s": 1.98, "r_error": "", "src_len": 1022, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 1048, "ts": 1791536274.0836635, "chk": "ok", "code_len": 1022}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6531, "tout": 571, "ts": 1791536298.3310952, "chk": null, "code_len": 1022}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791536264.6691537}
{"ep": "h2-00884", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.585, "speedup": 2.585, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 188.7, "calls": 2, "tokens_in": 7769, "tokens_out": 9440, "ctx_chars": 0, "t_end": 1791536319.840083, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 2.585, "lo95": 2.5826, "mean": 2.5794, "ev_only": 2.586, "ref_ms": 14.23638, "cand_ms": 5.50739, "n_trials": 88, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 64.86, "compile_s": 11.64, "r_error": "", "src_len": 5879, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6606, "ts": 1791536211.6298966, "chk": "ok", "code_len": 5879}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5586, "tout": 2834, "ts": 1791536254.802844, "chk": null, "code_len": 5879}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2183, "started": 1791536131.1146398}
{"ep": "h2-00916", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0092, "speedup": 1.0092, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 72.4, "calls": 1, "tokens_in": 6367, "tokens_out": 536, "ctx_chars": 10847, "t_end": 1791536320.4213102, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0092, "lo95": 1.0072, "mean": 1.0078, "ev_only": 1.0088, "ref_ms": 8.04156, "cand_ms": 7.96793, "n_trials": 144, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 48.78, "compile_s": 2.0, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 536, "ts": 1791536271.5561838, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791536248.0695188}
{"ep": "h2-00910", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.3992, "speedup": 1.3992, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 111.1, "calls": 2, "tokens_in": 4926, "tokens_out": 4722, "ctx_chars": 0, "t_end": 1791536322.6675773, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.3992, "lo95": 1.3983, "mean": 1.3991, "ev_only": 1.4009, "ref_ms": 15.59204, "cand_ms": 11.14393, "n_trials": 108, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9662, "r_wall_s": 44.41, "compile_s": 2.54, "r_error": "", "src_len": 2312, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 3226, "ts": 1791536240.0890174, "chk": "ok", "code_len": 2312}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3359, "tout": 1496, "ts": 1791536277.8855278, "chk": null, "code_len": 2312}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1567, "started": 1791536211.5812907}
{"ep": "h2-00914", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0019, "speedup": 1.0019, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 98.0, "calls": 2, "tokens_in": 4116, "tokens_out": 1922, "ctx_chars": 0, "t_end": 1791536323.240034, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0019, "lo95": 1.0013, "mean": 1.0007, "ev_only": 1.0011, "ref_ms": 7.97513, "cand_ms": 7.95965, "n_trials": 148, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 61.76, "compile_s": 2.02, "r_error": "", "src_len": 862, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 1393, "ts": 1791536238.7249224, "chk": "ok", "code_len": 862}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2492, "tout": 529, "ts": 1791536261.3156703, "chk": null, "code_len": 862}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791536225.284737}
{"ep": "h2-00930", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "incorrect", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 29.8, "calls": 1, "tokens_in": 1902, "tokens_out": 1110, "ctx_chars": 0, "t_end": 1791536325.4598901, "error": "output mismatch beyond atol=rtol=0.0001", "f_status": "incorrect", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "triton", "own_frac": 0.156, "r_wall_s": 19.08, "compile_s": 1.86, "r_error": "output mismatch beyond atol=rtol=0.0001", "src_len": 2176, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1902, "tout": 1110, "ts": 1791536306.1413786, "chk": null, "code_len": 2176}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1902, "started": 1791536295.6725233}
{"ep": "h2-00828", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 5.4956, "speedup": 5.4956, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 364.5, "calls": 2, "tokens_in": 6355, "tokens_out": 8334, "ctx_chars": 0, "t_end": 1791536327.967623, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 5.4956, "lo95": 5.494, "mean": 5.5632, "ev_only": 5.4958, "ref_ms": 7.79229, "cand_ms": 1.41792, "n_trials": 199, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 203.41, "compile_s": 14.17, "r_error": "", "src_len": 4933, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 5969, "ts": 1791536015.1997795, "chk": "ok", "code_len": 4933}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4630, "tout": 2365, "ts": 1791536123.8924289, "chk": null, "code_len": 4933}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1725, "started": 1791535963.4696882}
{"ep": "h2-00912", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7935, "speedup": 0.7935, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 119.1, "calls": 4, "tokens_in": 26350, "tokens_out": 1961, "ctx_chars": 8224, "t_end": 1791536338.1994846, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.7935, "lo95": 0.7901, "mean": 0.7926, "ev_only": 0.7934, "ref_ms": 60.35122, "cand_ms": 76.05383, "n_trials": 21, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 44.59, "compile_s": 2.13, "r_error": "", "src_len": 981, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 867, "ts": 1791536229.7236185, "chk": "runtime_error", "code_len": 989}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6485, "tout": 509, "ts": 1791536274.2994304, "chk": "ok", "code_len": 981}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7103, "tout": 107, "ts": 1791536289.0072887, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7234, "tout": 478, "ts": 1791536293.5675573, "chk": null, "code_len": 981}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["runtime_error", "ok", "no_code", null], "first_tin": 5528, "started": 1791536219.1146631}
{"ep": "h2-00911", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.6548, "speedup": 3.6548, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 126.1, "calls": 2, "tokens_in": 6574, "tokens_out": 6515, "ctx_chars": 0, "t_end": 1791536341.918368, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 3.6548, "lo95": 3.6011, "mean": 3.5318, "ev_only": 3.7701, "ref_ms": 1.80159, "cand_ms": 0.49294, "n_trials": 173, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 42.63, "compile_s": 13.14, "r_error": "", "src_len": 4600, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4079, "ts": 1791536250.4914708, "chk": "ok", "code_len": 4680}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4815, "tout": 2436, "ts": 1791536299.2416618, "chk": null, "code_len": 4600}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1759, "started": 1791536215.8244348}
{"ep": "h2-00928", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0009, "speedup": 1.0009, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 57.4, "calls": 1, "tokens_in": 6477, "tokens_out": 492, "ctx_chars": 10909, "t_end": 1791536343.852973, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0009, "lo95": 0.9996, "mean": 1.0035, "ev_only": 1.0019, "ref_ms": 8.06484, "cand_ms": 8.05753, "n_trials": 158, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 52.26, "compile_s": 2.05, "r_error": "", "src_len": 949, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 492, "ts": 1791536291.5166101, "chk": null, "code_len": 949}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791536286.4966302}
{"ep": "h2-00913", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.2893, "speedup": 0.2893, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 124.3, "calls": 4, "tokens_in": 37516, "tokens_out": 8109, "ctx_chars": 6513, "t_end": 1791536344.3400855, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.2893, "lo95": 0.2873, "mean": 0.3059, "ev_only": 0.2892, "ref_ms": 7.50025, "cand_ms": 25.92111, "n_trials": 61, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0507, "r_wall_s": 28.29, "compile_s": 2.42, "r_error": "", "src_len": 5521, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2955, "ts": 1791536244.768947, "chk": "incorrect", "code_len": 6848}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8732, "tout": 2700, "ts": 1791536281.1021569, "chk": "ok", "code_len": 5622}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11544, "tout": 27, "ts": 1791536296.2289736, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11595, "tout": 2427, "ts": 1791536316.0064454, "chk": null, "code_len": 5521}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791536220.0829022}
{"ep": "h2-00898", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9474, "speedup": 0.9474, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 176.8, "calls": 2, "tokens_in": 15494, "tokens_out": 11785, "ctx_chars": 6513, "t_end": 1791536345.9155505, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9474, "lo95": 0.9369, "mean": 0.9078, "ev_only": 0.9481, "ref_ms": 7.19709, "cand_ms": 7.59629, "n_trials": 204, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.3002, "r_wall_s": 54.4, "compile_s": 2.07, "r_error": "", "src_len": 8210, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 7382, "ts": 1791536232.1211994, "chk": "ok", "code_len": 8280}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9847, "tout": 4403, "ts": 1791536291.4909906, "chk": null, "code_len": 8210}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791536169.6898017}
{"ep": "h2-00915", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.3319, "speedup": 0.3319, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 110.7, "calls": 3, "tokens_in": 22567, "tokens_out": 5914, "ctx_chars": 6513, "t_end": 1791536347.3639157, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.3319, "lo95": 0.3302, "mean": 0.3411, "ev_only": 0.3303, "ref_ms": 3.41614, "cand_ms": 10.29396, "n_trials": 87, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0264, "r_wall_s": 53.1, "compile_s": 2.12, "r_error": "", "src_len": 6091, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2973, "ts": 1791536261.454677, "chk": "ok", "code_len": 6091}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8531, "tout": 35, "ts": 1791536271.3897421, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8590, "tout": 2906, "ts": 1791536294.2372193, "chk": null, "code_len": 6091}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791536237.5944}
{"ep": "h2-00798", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0056, "speedup": 1.0056, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 438.9, "calls": 4, "tokens_in": 21012, "tokens_out": 6127, "ctx_chars": 0, "t_end": 1791536352.4277256, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0056, "lo95": 1.0049, "mean": 1.0407, "ev_only": 1.0057, "ref_ms": 30.43623, "cand_ms": 30.26756, "n_trials": 47, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.042, "r_wall_s": 243.34, "compile_s": 12.35, "r_error": "", "src_len": 4427, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1623, "tout": 2127, "ts": 1791535932.133794, "chk": "compile_error", "code_len": 4496}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5043, "tout": 1984, "ts": 1791535995.4417312, "chk": "ok", "code_len": 4427}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7139, "tout": 44, "ts": 1791536092.7455335, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7207, "tout": 1972, "ts": 1791536108.6167705, "chk": null, "code_len": 4427}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 1623, "started": 1791535913.5052853}
{"ep": "h2-00926", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0156, "speedup": 1.0156, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 72.7, "calls": 2, "tokens_in": 15998, "tokens_out": 3568, "ctx_chars": 11065, "t_end": 1791536357.00084, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0156, "lo95": 1.0064, "mean": 1.0247, "ev_only": 1.02, "ref_ms": 1.86579, "cand_ms": 1.83714, "n_trials": 232, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2794, "r_wall_s": 28.17, "compile_s": 1.94, "r_error": "", "src_len": 1917, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 2632, "ts": 1791536310.6585748, "chk": "ok", "code_len": 1917}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9371, "tout": 936, "ts": 1791536328.5814095, "chk": null, "code_len": 1917}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791536284.300904}
{"ep": "h2-00893", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.3159, "speedup": 1.3159, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 206.6, "calls": 2, "tokens_in": 10207, "tokens_out": 16856, "ctx_chars": 0, "t_end": 1791536357.1512072, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.3159, "lo95": 1.3151, "mean": 1.3452, "ev_only": 1.3164, "ref_ms": 15.72185, "cand_ms": 11.94761, "n_trials": 80, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.5272, "r_wall_s": 37.09, "compile_s": 11.7, "r_error": "", "src_len": 10794, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2161, "tout": 11593, "ts": 1791536248.9045792, "chk": "ok", "code_len": 10794}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8046, "tout": 5263, "ts": 1791536319.5157168, "chk": null, "code_len": 10794}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2161, "started": 1791536150.554211}
{"ep": "h2-00920", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0016, "speedup": 1.0016, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 99.5, "calls": 2, "tokens_in": 4155, "tokens_out": 1824, "ctx_chars": 0, "t_end": 1791536357.8155758, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0016, "lo95": 0.9995, "mean": 0.9982, "ev_only": 1.0028, "ref_ms": 8.01524, "cand_ms": 8.00253, "n_trials": 172, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 32.07, "compile_s": 1.92, "r_error": "", "src_len": 908, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 1253, "ts": 1791536283.5204744, "chk": "ok", "code_len": 908}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2531, "tout": 571, "ts": 1791536325.4507298, "chk": null, "code_len": 908}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791536258.331888}
{"ep": "h2-00841", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.6524, "speedup": 0.6524, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 351.3, "calls": 2, "tokens_in": 12479, "tokens_out": 20027, "ctx_chars": 0, "t_end": 1791536358.7236955, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.6524, "lo95": 0.6496, "mean": 0.685, "ev_only": 0.6503, "ref_ms": 7.46782, "cand_ms": 11.44743, "n_trials": 50, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.9992, "r_wall_s": 54.5, "compile_s": 1.95, "r_error": "", "src_len": 11910, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 13628, "ts": 1791536122.1825805, "chk": "ok", "code_len": 11910}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9535, "tout": 6399, "ts": 1791536304.1784143, "chk": null, "code_len": 11910}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791536007.376121}
{"ep": "h2-00870", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.497, "speedup": 1.497, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 265.5, "calls": 2, "tokens_in": 5174, "tokens_out": 4287, "ctx_chars": 0, "t_end": 1791536359.6018755, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.497, "lo95": 1.4962, "mean": 1.4964, "ev_only": 1.4997, "ref_ms": 7.98528, "cand_ms": 5.33428, "n_trials": 189, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 30.89, "compile_s": 1.89, "r_error": "", "src_len": 2733, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 2798, "ts": 1791536148.8327868, "chk": "ok", "code_len": 2733}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3573, "tout": 1489, "ts": 1791536328.6423423, "chk": null, "code_len": 2733}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791536094.0663116}
{"ep": "h2-00829", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.9706, "speedup": 4.9706, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 392.8, "calls": 3, "tokens_in": 13449, "tokens_out": 9732, "ctx_chars": 0, "t_end": 1791536364.7964063, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 4.9706, "lo95": 4.9442, "mean": 4.8479, "ev_only": 5.1015, "ref_ms": 1.85035, "cand_ms": 0.37226, "n_trials": 194, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 84.91, "compile_s": 18.21, "r_error": "", "src_len": 4327, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 5104, "ts": 1791536015.674998, "chk": "no_custom_kernel", "code_len": 4349}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4662, "tout": 2402, "ts": 1791536124.665794, "chk": "ok", "code_len": 4327}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7028, "tout": 2226, "ts": 1791536279.7617366, "chk": null, "code_len": 4327}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1759, "started": 1791535972.0004785}
{"ep": "h2-00909", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8788, "speedup": 0.8788, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 163.1, "calls": 2, "tokens_in": 5822, "tokens_out": 5197, "ctx_chars": 0, "t_end": 1791536367.0749385, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.8788, "lo95": 0.8771, "mean": 0.8787, "ev_only": 0.8785, "ref_ms": 8.16639, "cand_ms": 9.29252, "n_trials": 71, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 76.54, "compile_s": 3.45, "r_error": "", "src_len": 3030, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 3500, "ts": 1791536235.8389711, "chk": "ok", "code_len": 3030}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4070, "tout": 1697, "ts": 1791536290.380144, "chk": null, "code_len": 3030}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1752, "started": 1791536204.023161}
{"ep": "h2-00863", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.1697, "speedup": 1.1697, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 294.1, "calls": 1, "tokens_in": 1757, "tokens_out": 3981, "ctx_chars": 0, "t_end": 1791536372.4361997, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.1697, "lo95": 1.1671, "mean": 1.1662, "ev_only": 1.1741, "ref_ms": 1.81006, "cand_ms": 1.54752, "n_trials": 188, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1657, "r_wall_s": 254.41, "compile_s": 57.84, "r_error": "", "src_len": 2552, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1757, "tout": 3981, "ts": 1791536117.9957106, "chk": null, "code_len": 2552}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1757, "started": 1791536078.3352296}
{"ep": "h2-00902", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4814, "speedup": 1.4814, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 184.6, "calls": 2, "tokens_in": 6829, "tokens_out": 7044, "ctx_chars": 0, "t_end": 1791536372.9026377, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.4814, "lo95": 1.4798, "mean": 1.4812, "ev_only": 1.4808, "ref_ms": 31.37829, "cand_ms": 21.1809, "n_trials": 48, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9997, "r_wall_s": 88.28, "compile_s": 2.77, "r_error": "", "src_len": 3916, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 4937, "ts": 1791536240.035184, "chk": "ok", "code_len": 3960}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4751, "tout": 2107, "ts": 1791536283.312246, "chk": null, "code_len": 3916}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791536188.2693858}
{"ep": "h2-00866", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.8559, "speedup": 0.8559, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 286.5, "calls": 3, "tokens_in": 15775, "tokens_out": 13182, "ctx_chars": 0, "t_end": 1791536377.1710742, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8559, "lo95": 0.8556, "mean": 0.8559, "ev_only": 0.8561, "ref_ms": 30.0982, "cand_ms": 35.16454, "n_trials": 35, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.9988, "r_wall_s": 66.82, "compile_s": 2.02, "r_error": "", "src_len": 6086, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 6017, "ts": 1791536142.0681129, "chk": "forbidden", "code_len": 5977}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5364, "tout": 3441, "ts": 1791536165.3925571, "chk": "ok", "code_len": 6086}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8813, "tout": 3724, "ts": 1791536309.8841617, "chk": null, "code_len": 6086}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", null], "first_tin": 1598, "started": 1791536090.716493}
{"ep": "h2-00881", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.3317, "speedup": 2.3317, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 251.4, "calls": 3, "tokens_in": 21009, "tokens_out": 3951, "ctx_chars": 8529, "t_end": 1791536378.3481576, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 2.3317, "lo95": 2.329, "mean": 2.3456, "ev_only": 2.3394, "ref_ms": 8.51683, "cand_ms": 3.6526, "n_trials": 176, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4274, "r_wall_s": 89.66, "compile_s": 3.74, "r_error": "", "src_len": 3463, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2111, "ts": 1791536146.762341, "chk": "ok", "code_len": 3499}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7718, "tout": 54, "ts": 1791536274.1003346, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7796, "tout": 1786, "ts": 1791536288.4856665, "chk": null, "code_len": 3463}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5495, "started": 1791536126.9653134}
{"ep": "h2-00931", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3338, "speedup": 1.3338, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 83.7, "calls": 1, "tokens_in": 5998, "tokens_out": 732, "ctx_chars": 9970, "t_end": 1791536380.8935091, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.3338, "lo95": 1.333, "mean": 1.3338, "ev_only": 1.3353, "ref_ms": 7.82158, "cand_ms": 5.86433, "n_trials": 209, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1109, "r_wall_s": 75.45, "compile_s": 1.98, "r_error": "", "src_len": 1457, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 732, "ts": 1791536304.6016777, "chk": null, "code_len": 1457}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791536297.7376356}
{"ep": "h2-00874", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.4543, "speedup": 1.4543, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 280.4, "calls": 2, "tokens_in": 11177, "tokens_out": 4959, "ctx_chars": 6513, "t_end": 1791536387.4153666, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.4543, "lo95": 1.4533, "mean": 1.4526, "ev_only": 1.4538, "ref_ms": 15.50164, "cand_ms": 10.65886, "n_trials": 101, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1654, "r_wall_s": 106.02, "compile_s": 3.01, "r_error": "", "src_len": 3178, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3514, "ts": 1791536138.439137, "chk": "ok", "code_len": 3178}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6570, "tout": 1445, "ts": 1791536280.3802338, "chk": null, "code_len": 3178}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791536106.967836}
{"ep": "h2-00901", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5465, "speedup": 1.5465, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 201.5, "calls": 2, "tokens_in": 6440, "tokens_out": 5776, "ctx_chars": 0, "t_end": 1791536388.9020283, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.5465, "lo95": 1.5459, "mean": 1.5463, "ev_only": 1.5471, "ref_ms": 31.37669, "cand_ms": 20.28824, "n_trials": 43, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9554, "r_wall_s": 106.43, "compile_s": 3.4, "r_error": "", "src_len": 3327, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 3967, "ts": 1791536240.9027703, "chk": "ok", "code_len": 3327}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4362, "tout": 1809, "ts": 1791536281.6169186, "chk": null, "code_len": 3327}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791536187.3833234}
{"ep": "h2-00921", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0023, "speedup": 1.0023, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 137.5, "calls": 2, "tokens_in": 4168, "tokens_out": 2255, "ctx_chars": 0, "t_end": 1791536399.044929, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0023, "lo95": 0.9989, "mean": 1.006, "ev_only": 0.998, "ref_ms": 8.05557, "cand_ms": 8.03702, "n_trials": 85, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0023, "r_wall_s": 48.75, "compile_s": 2.18, "r_error": "", "src_len": 1020, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1656, "ts": 1791536313.699994, "chk": "ok", "code_len": 1020}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2665, "tout": 599, "ts": 1791536349.9505131, "chk": null, "code_len": 1020}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791536261.5350327}
{"ep": "h2-00871", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9162, "speedup": 0.9162, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 305.0, "calls": 3, "tokens_in": 24026, "tokens_out": 2071, "ctx_chars": 11539, "t_end": 1791536399.749599, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9162, "lo95": 0.9148, "mean": 0.9109, "ev_only": 0.9195, "ref_ms": 7.81955, "cand_ms": 8.53454, "n_trials": 162, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0767, "r_wall_s": 128.52, "compile_s": 1.95, "r_error": "", "src_len": 1772, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1180, "ts": 1791536124.4471967, "chk": "ok", "code_len": 1772}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8402, "tout": 85, "ts": 1791536263.5717754, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8511, "tout": 806, "ts": 1791536270.5236645, "chk": null, "code_len": 1772}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791536094.7923553}
{"ep": "h2-00943", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0059, "speedup": 1.0059, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 72.4, "calls": 2, "tokens_in": 13740, "tokens_out": 1264, "ctx_chars": 10909, "t_end": 1791536400.338859, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0059, "lo95": 1.0029, "mean": 1.0081, "ev_only": 1.0081, "ref_ms": 8.0081, "cand_ms": 7.96105, "n_trials": 162, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 48.01, "compile_s": 2.99, "r_error": "", "src_len": 938, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 673, "ts": 1791536334.2580223, "chk": "ok", "code_len": 938}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7261, "tout": 591, "ts": 1791536352.2453382, "chk": null, "code_len": 938}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791536327.9678848}
{"ep": "h2-00794", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.8847, "speedup": 1.8847, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 492.4, "calls": 3, "tokens_in": 17424, "tokens_out": 14458, "ctx_chars": 0, "t_end": 1791536402.124009, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.8847, "lo95": 1.884, "mean": 1.8838, "ev_only": 1.8822, "ref_ms": 7.75845, "cand_ms": 4.11648, "n_trials": 173, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 240.14, "compile_s": 20.01, "r_error": "", "src_len": 6278, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 7800, "ts": 1791535978.09768, "chk": "no_custom_kernel", "code_len": 6301}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5968, "tout": 3510, "ts": 1791536028.0151768, "chk": "ok", "code_len": 6278}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9297, "tout": 3148, "ts": 1791536161.3368719, "chk": null, "code_len": 6278}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791535909.7665863}
{"ep": "h2-00932", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.535, "speedup": 0.535, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 93.0, "calls": 1, "tokens_in": 1565, "tokens_out": 1697, "ctx_chars": 0, "t_end": 1791536405.468461, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.535, "lo95": 0.5348, "mean": 0.5351, "ev_only": 0.535, "ref_ms": 15.51539, "cand_ms": 29.00065, "n_trials": 47, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 73.01, "compile_s": 2.05, "r_error": "", "src_len": 2905, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1565, "tout": 1697, "ts": 1791536331.9180713, "chk": null, "code_len": 2905}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1565, "started": 1791536312.455103}
{"ep": "h2-00865", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0664, "speedup": 1.0664, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 322.7, "calls": 2, "tokens_in": 14483, "tokens_out": 8255, "ctx_chars": 6513, "t_end": 1791536406.408163, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0664, "lo95": 0.8295, "mean": 0.9358, "ev_only": 1.0678, "ref_ms": 4.55592, "cand_ms": 4.27242, "n_trials": 175, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1058, "r_wall_s": 40.61, "compile_s": 3.1, "r_error": "", "src_len": 6670, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 5047, "ts": 1791536125.2152553, "chk": "ok", "code_len": 6670}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9035, "tout": 3208, "ts": 1791536365.776139, "chk": null, "code_len": 6670}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791536083.7528734}
{"ep": "h2-00924", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9961, "speedup": 0.9961, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 129.1, "calls": 2, "tokens_in": 13389, "tokens_out": 981, "ctx_chars": 10847, "t_end": 1791536407.092943, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9961, "lo95": 0.9916, "mean": 1.0007, "ev_only": 0.9989, "ref_ms": 8.09904, "cand_ms": 8.13041, "n_trials": 148, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0038, "r_wall_s": 74.4, "compile_s": 2.01, "r_error": "", "src_len": 823, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 542, "ts": 1791536286.0894265, "chk": "ok", "code_len": 823}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7020, "tout": 439, "ts": 1791536332.5923185, "chk": null, "code_len": 823}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791536278.0030994}
{"ep": "h2-00948", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0058, "speedup": 1.0058, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 61.3, "calls": 1, "tokens_in": 6477, "tokens_out": 522, "ctx_chars": 10909, "t_end": 1791536407.2371945, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0058, "lo95": 1.0038, "mean": 1.0071, "ev_only": 1.006, "ref_ms": 8.03584, "cand_ms": 7.98916, "n_trials": 181, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 55.73, "compile_s": 1.87, "r_error": "", "src_len": 970, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 522, "ts": 1791536351.43295, "chk": null, "code_len": 970}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791536345.9158113}
{"ep": "h2-00900", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.5316, "speedup": 1.5316, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 227.3, "calls": 3, "tokens_in": 10773, "tokens_out": 4072, "ctx_chars": 0, "t_end": 1791536408.6300154, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.5316, "lo95": 1.5267, "mean": 1.5325, "ev_only": 1.5308, "ref_ms": 8.58422, "cand_ms": 5.60467, "n_trials": 158, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.6337, "r_wall_s": 128.26, "compile_s": 3.18, "r_error": "", "src_len": 2843, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1818, "tout": 2459, "ts": 1791536239.4369187, "chk": "ok", "code_len": 2934}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4389, "tout": 153, "ts": 1791536268.3431659, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 4566, "tout": 1460, "ts": 1791536280.3127096, "chk": null, "code_len": 2843}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1818, "started": 1791536181.339163}
{"ep": "h2-00853", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0541, "speedup": 1.0541, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 358.8, "calls": 2, "tokens_in": 4687, "tokens_out": 2852, "ctx_chars": 0, "t_end": 1791536411.503396, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0541, "lo95": 1.0531, "mean": 1.0516, "ev_only": 1.0555, "ref_ms": 25.02114, "cand_ms": 23.73709, "n_trials": 58, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0319, "r_wall_s": 144.01, "compile_s": 1.94, "r_error": "", "src_len": 1351, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 2091, "ts": 1791536129.084724, "chk": "ok", "code_len": 1351}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3019, "tout": 761, "ts": 1791536266.7244487, "chk": null, "code_len": 1351}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791536052.6809702}
{"ep": "h2-00812", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.4665, "speedup": 2.4665, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 472.8, "calls": 3, "tokens_in": 16380, "tokens_out": 12493, "ctx_chars": 0, "t_end": 1791536412.2231455, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 2.4665, "lo95": 2.4642, "mean": 2.5145, "ev_only": 2.4712, "ref_ms": 14.20806, "cand_ms": 5.7605, "n_trials": 103, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 258.7, "compile_s": 20.85, "r_error": "", "src_len": 5802, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6325, "ts": 1791535994.73038, "chk": "no_custom_kernel", "code_len": 5823}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5583, "tout": 3336, "ts": 1791536077.3027914, "chk": "ok", "code_len": 5802}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8614, "tout": 2832, "ts": 1791536153.4324372, "chk": null, "code_len": 5802}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2183, "started": 1791535939.4197152}
{"ep": "h2-00882", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9249, "speedup": 0.9249, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 288.9, "calls": 3, "tokens_in": 24136, "tokens_out": 2449, "ctx_chars": 11539, "t_end": 1791536416.0893297, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9249, "lo95": 0.9244, "mean": 0.9268, "ev_only": 0.9234, "ref_ms": 7.76017, "cand_ms": 8.39042, "n_trials": 186, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0757, "r_wall_s": 58.89, "compile_s": 2.15, "r_error": "", "src_len": 2034, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1078, "ts": 1791536156.6802466, "chk": "ok", "code_len": 2034}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8300, "tout": 399, "ts": 1791536339.5111847, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8723, "tout": 972, "ts": 1791536356.1990519, "chk": null, "code_len": 2034}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791536127.2184546}
{"ep": "h2-00938", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0038, "speedup": 1.0038, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.9, "calls": 2, "tokens_in": 11726, "tokens_out": 1461, "ctx_chars": 9028, "t_end": 1791536417.7850947, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0038, "lo95": 1.0029, "mean": 1.0082, "ev_only": 1.0042, "ref_ms": 8.01692, "cand_ms": 7.98645, "n_trials": 184, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0071, "r_wall_s": 57.26, "compile_s": 2.16, "r_error": "", "src_len": 998, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 872, "ts": 1791536327.4692354, "chk": "ok", "code_len": 998}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6355, "tout": 589, "ts": 1791536360.390877, "chk": null, "code_len": 998}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791536319.8403156}
{"ep": "h2-00878", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0083, "speedup": 1.0083, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 305.4, "calls": 2, "tokens_in": 15091, "tokens_out": 4960, "ctx_chars": 10823, "t_end": 1791536418.6386993, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0083, "lo95": 1.0079, "mean": 1.0087, "ev_only": 1.0079, "ref_ms": 30.12231, "cand_ms": 29.87355, "n_trials": 21, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0074, "r_wall_s": 81.59, "compile_s": 2.0, "r_error": "", "src_len": 3628, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 3143, "ts": 1791536139.9310062, "chk": "ok", "code_len": 3628}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8736, "tout": 1817, "ts": 1791536336.448229, "chk": null, "code_len": 3628}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791536113.1929307}
{"ep": "h2-00903", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0246, "speedup": 1.0246, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 230.3, "calls": 2, "tokens_in": 6830, "tokens_out": 7992, "ctx_chars": 0, "t_end": 1791536419.014412, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0246, "lo95": 1.0231, "mean": 1.0251, "ev_only": 1.0244, "ref_ms": 30.62148, "cand_ms": 29.88501, "n_trials": 41, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0163, "r_wall_s": 123.53, "compile_s": 14.72, "r_error": "", "src_len": 6449, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 4974, "ts": 1791536240.4140847, "chk": "ok", "code_len": 6449}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5205, "tout": 3018, "ts": 1791536294.9895027, "chk": null, "code_len": 6449}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791536188.7563593}
{"ep": "h2-00935", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9765, "speedup": 0.9765, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 110.5, "calls": 1, "tokens_in": 7032, "tokens_out": 1137, "ctx_chars": 11539, "t_end": 1791536427.1294603, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9765, "lo95": 0.976, "mean": 0.9754, "ev_only": 0.9764, "ref_ms": 31.40457, "cand_ms": 32.16136, "n_trials": 41, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0386, "r_wall_s": 95.69, "compile_s": 1.93, "r_error": "", "src_len": 1798, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1137, "ts": 1791536330.587494, "chk": null, "code_len": 1798}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7032, "started": 1791536316.5972137}
{"ep": "h2-00880", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0044, "speedup": 1.0044, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 307.2, "calls": 3, "tokens_in": 20017, "tokens_out": 2708, "ctx_chars": 9266, "t_end": 1791536432.0610619, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0044, "lo95": 1.0026, "mean": 1.0069, "ev_only": 1.005, "ref_ms": 30.59235, "cand_ms": 30.45698, "n_trials": 20, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0441, "r_wall_s": 93.13, "compile_s": 2.05, "r_error": "", "src_len": 2345, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1409, "ts": 1791536154.9457507, "chk": "ok", "code_len": 2345}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7135, "tout": 109, "ts": 1791536328.7172267, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7268, "tout": 1190, "ts": 1791536338.5047119, "chk": null, "code_len": 2345}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791536124.8908396}
{"ep": "h2-00877", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0064, "speedup": 1.0064, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 327.9, "calls": 2, "tokens_in": 16314, "tokens_out": 2720, "ctx_chars": 12117, "t_end": 1791536438.6497784, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0064, "lo95": 1.0055, "mean": 1.0128, "ev_only": 1.0065, "ref_ms": 42.43176, "cand_ms": 42.16036, "n_trials": 33, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 78.66, "compile_s": 2.1, "r_error": "", "src_len": 2112, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1638, "ts": 1791536124.4261339, "chk": "ok", "code_len": 2112}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8915, "tout": 1082, "ts": 1791536359.2096105, "chk": null, "code_len": 2112}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791536110.7068653}
{"ep": "h2-00929", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.03, "speedup": 1.03, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 149.1, "calls": 2, "tokens_in": 12677, "tokens_out": 1489, "ctx_chars": 9656, "t_end": 1791536439.1382542, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.03, "lo95": 1.0271, "mean": 1.1009, "ev_only": 1.0293, "ref_ms": 25.02494, "cand_ms": 24.29641, "n_trials": 23, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0226, "r_wall_s": 118.44, "compile_s": 2.23, "r_error": "", "src_len": 1031, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5834, "tout": 897, "ts": 1791536297.5250113, "chk": "ok", "code_len": 1031}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6843, "tout": 592, "ts": 1791536319.9140017, "chk": null, "code_len": 1031}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5834, "started": 1791536289.9990222}
{"ep": "h2-00873", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 6.5663, "speedup": 6.5663, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 343.7, "calls": 3, "tokens_in": 18062, "tokens_out": 20413, "ctx_chars": 0, "t_end": 1791536443.7927988, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 6.5663, "lo95": 6.5424, "mean": 6.7834, "ev_only": 6.5588, "ref_ms": 32.62662, "cand_ms": 4.96882, "n_trials": 43, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9759, "r_wall_s": 94.94, "compile_s": 14.08, "r_error": "", "src_len": 7340, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 12716, "ts": 1791536218.6908474, "chk": "no_custom_kernel", "code_len": 7344}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6116, "tout": 4049, "ts": 1791536291.365114, "chk": "ok", "code_len": 7340}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9980, "tout": 3648, "ts": 1791536348.8025627, "chk": null, "code_len": 7340}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791536100.0768974}
{"ep": "h2-00872", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0085, "speedup": 1.0085, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 356.0, "calls": 2, "tokens_in": 15306, "tokens_out": 6222, "ctx_chars": 10823, "t_end": 1791536453.982662, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0085, "lo95": 1.0082, "mean": 1.0089, "ev_only": 1.0084, "ref_ms": 30.10213, "cand_ms": 29.84735, "n_trials": 51, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0197, "r_wall_s": 114.18, "compile_s": 2.0, "r_error": "", "src_len": 3903, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 4353, "ts": 1791536155.0812151, "chk": "ok", "code_len": 3903}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8951, "tout": 1869, "ts": 1791536339.3447924, "chk": null, "code_len": 3903}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791536097.978481}
{"ep": "h2-00939", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.4325, "speedup": 0.4325, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 136.3, "calls": 3, "tokens_in": 22561, "tokens_out": 5884, "ctx_chars": 6513, "t_end": 1791536456.7693167, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.4325, "lo95": 0.3662, "mean": 0.4008, "ev_only": 0.4316, "ref_ms": 4.96053, "cand_ms": 11.46884, "n_trials": 130, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0248, "r_wall_s": 79.22, "compile_s": 3.08, "r_error": "", "src_len": 5887, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 2954, "ts": 1791536344.2498984, "chk": "ok", "code_len": 5923}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8512, "tout": 67, "ts": 1791536355.1283598, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8603, "tout": 2863, "ts": 1791536377.5196269, "chk": null, "code_len": 5887}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5446, "started": 1791536320.4215455}
{"ep": "h2-00979", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8085, "speedup": 0.8085, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 38.0, "calls": 3, "tokens_in": 18532, "tokens_out": 1432, "ctx_chars": 8224, "t_end": 1791536457.059384, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.8085, "lo95": 0.8051, "mean": 0.8077, "ev_only": 0.8083, "ref_ms": 31.93018, "cand_ms": 39.49314, "n_trials": 35, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 13.31, "compile_s": 1.87, "r_error": "", "src_len": 1060, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 843, "ts": 1791536428.7716596, "chk": "ok", "code_len": 1060}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6460, "tout": 80, "ts": 1791536438.8074627, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6564, "tout": 509, "ts": 1791536443.7097833, "chk": null, "code_len": 1060}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791536419.0146832}
{"ep": "h2-00919", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0475, "speedup": 1.0475, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 202.2, "calls": 2, "tokens_in": 5542, "tokens_out": 4945, "ctx_chars": 0, "t_end": 1791536459.8099961, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0475, "lo95": 1.0462, "mean": 1.0476, "ev_only": 1.0477, "ref_ms": 27.48008, "cand_ms": 26.23358, "n_trials": 43, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0163, "r_wall_s": 25.65, "compile_s": 12.48, "r_error": "", "src_len": 3319, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 3264, "ts": 1791536289.59745, "chk": "ok", "code_len": 3319}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3812, "tout": 1681, "ts": 1791536434.1182263, "chk": null, "code_len": 3319}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791536257.6517873}
{"ep": "h2-00941", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0433, "speedup": 1.0433, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 140.9, "calls": 3, "tokens_in": 18614, "tokens_out": 3742, "ctx_chars": 6513, "t_end": 1791536464.0913177, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0433, "lo95": 1.0417, "mean": 1.0493, "ev_only": 1.0457, "ref_ms": 15.77476, "cand_ms": 15.11977, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1441, "r_wall_s": 99.21, "compile_s": 3.28, "r_error": "", "src_len": 3775, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1876, "ts": 1791536338.973318, "chk": "ok", "code_len": 3775}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6850, "tout": 28, "ts": 1791536350.1199539, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6902, "tout": 1838, "ts": 1791536364.6285598, "chk": null, "code_len": 3775}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791536323.240252}
{"ep": "h2-00960", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.002, "speedup": 1.002, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 88.9, "calls": 1, "tokens_in": 6477, "tokens_out": 521, "ctx_chars": 10909, "t_end": 1791536466.1126935, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.002, "lo95": 0.9999, "mean": 1.0204, "ev_only": 1.003, "ref_ms": 8.12603, "cand_ms": 8.10964, "n_trials": 142, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 55.13, "compile_s": 5.05, "r_error": "", "src_len": 976, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 521, "ts": 1791536410.880889, "chk": null, "code_len": 976}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791536377.1713033}
{"ep": "h2-00974", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 55.4, "calls": 1, "tokens_in": 1599, "tokens_out": 936, "ctx_chars": 0, "t_end": 1791536466.8696136, "error": "RuntimeError: Error building extension 'div_gelu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=div_gelu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 46.1, "compile_s": 46.1, "r_error": "RuntimeError: Error building extension 'div_gelu_mod': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=div_gelu_mod -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "src_len": 1975, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1599, "tout": 936, "ts": 1791536420.5157769, "chk": null, "code_len": 1975}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1599, "started": 1791536411.503647}
{"ep": "h2-00965", "task": "L2_094", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0034, "speedup": 1.0034, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 68.1, "calls": 2, "tokens_in": 6145, "tokens_out": 3516, "ctx_chars": 0, "t_end": 1791536467.1178868, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0034, "lo95": 1.0025, "mean": 1.0021, "ev_only": 1.0041, "ref_ms": 8.36376, "cand_ms": 8.33553, "n_trials": 151, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0429, "r_wall_s": 17.85, "compile_s": 2.19, "r_error": "", "src_len": 2087, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1764, "tout": 2505, "ts": 1791536425.1606517, "chk": "ok", "code_len": 2087}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4381, "tout": 1011, "ts": 1791536449.1926599, "chk": null, "code_len": 2087}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1764, "started": 1791536399.0451815}
{"ep": "h2-00947", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0068, "speedup": 1.0068, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 123.1, "calls": 2, "tokens_in": 12812, "tokens_out": 2683, "ctx_chars": 9266, "t_end": 1791536467.4841025, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0068, "lo95": 1.0064, "mean": 1.0067, "ev_only": 1.0066, "ref_ms": 30.47415, "cand_ms": 30.26795, "n_trials": 44, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0409, "r_wall_s": 78.02, "compile_s": 2.08, "r_error": "", "src_len": 2282, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1472, "ts": 1791536358.3536599, "chk": "ok", "code_len": 2282}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7198, "tout": 1211, "ts": 1791536389.0315397, "chk": null, "code_len": 2282}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5614, "started": 1791536344.3403213}
{"ep": "h2-00983", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.8479, "speedup": 0.8479, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 30.0, "calls": 1, "tokens_in": 7030, "tokens_out": 1219, "ctx_chars": 11175, "t_end": 1791536469.1848364, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8479, "lo95": 0.8466, "mean": 0.8472, "ev_only": 0.8481, "ref_ms": 14.29818, "cand_ms": 16.86364, "n_trials": 74, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 18.56, "compile_s": 2.12, "r_error": "", "src_len": 2537, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1219, "ts": 1791536450.3017347, "chk": null, "code_len": 2537}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791536439.1384869}
{"ep": "h2-00969", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0199, "speedup": 1.0199, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 66.4, "calls": 3, "tokens_in": 18307, "tokens_out": 4129, "ctx_chars": 5026, "t_end": 1791536471.9157481, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0199, "lo95": 1.0193, "mean": 1.0677, "ev_only": 1.0198, "ref_ms": 8.14837, "cand_ms": 7.98956, "n_trials": 146, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0092, "r_wall_s": 19.47, "compile_s": 1.87, "r_error": "", "src_len": 2051, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 3004, "ts": 1791536431.9283028, "chk": "ok", "code_len": 2051}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7109, "tout": 72, "ts": 1791536443.3942025, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7205, "tout": 1053, "ts": 1791536452.288771, "chk": null, "code_len": 2051}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 3993, "started": 1791536405.4686956}
{"ep": "h2-00963", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 93.7, "calls": 1, "tokens_in": 2743, "tokens_out": 3035, "ctx_chars": 0, "t_end": 1791536481.163785, "error": "RuntimeError: Error building extension 'resnet_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=resnet_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 67.42, "compile_s": 67.35, "r_error": "RuntimeError: Error building extension 'resnet_fused': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=resnet_fused -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/lo", "src_len": 6617, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2743, "tout": 3035, "ts": 1791536413.7140696, "chk": null, "code_len": 6617}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2743, "started": 1791536387.4156125}
{"ep": "h2-00888", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.5523, "speedup": 0.5523, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 335.4, "calls": 3, "tokens_in": 20119, "tokens_out": 3191, "ctx_chars": 9033, "t_end": 1791536481.5079257, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.5523, "lo95": 0.5518, "mean": 0.5516, "ev_only": 0.5519, "ref_ms": 15.63126, "cand_ms": 28.3027, "n_trials": 42, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 59.78, "compile_s": 1.93, "r_error": "", "src_len": 2651, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 1633, "ts": 1791536162.610185, "chk": "ok", "code_len": 2651}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7230, "tout": 147, "ts": 1791536409.6770763, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7401, "tout": 1411, "ts": 1791536421.4156404, "chk": null, "code_len": 2651}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5488, "started": 1791536146.058172}
{"ep": "h2-00977", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0168, "speedup": 1.0168, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 64.5, "calls": 2, "tokens_in": 15472, "tokens_out": 3601, "ctx_chars": 11823, "t_end": 1791536482.2925982, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0168, "lo95": 1.0149, "mean": 1.0273, "ev_only": 1.0167, "ref_ms": 27.75191, "cand_ms": 27.29405, "n_trials": 38, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0121, "r_wall_s": 23.29, "compile_s": 2.06, "r_error": "", "src_len": 2201, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6923, "tout": 2338, "ts": 1791536438.7246864, "chk": "ok", "code_len": 2201}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8549, "tout": 1263, "ts": 1791536458.9698803, "chk": null, "code_len": 2201}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6923, "started": 1791536417.7853315}
{"ep": "h2-00946", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0049, "speedup": 1.0049, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.4, "calls": 2, "tokens_in": 16178, "tokens_out": 2514, "ctx_chars": 12117, "t_end": 1791536498.285288, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0049, "lo95": 1.0036, "mean": 1.0054, "ev_only": 1.0048, "ref_ms": 42.36674, "cand_ms": 42.16034, "n_trials": 33, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 60.18, "compile_s": 1.94, "r_error": "", "src_len": 2054, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1447, "ts": 1791536355.7591026, "chk": "ok", "code_len": 2054}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8779, "tout": 1067, "ts": 1791536437.2133164, "chk": null, "code_len": 2054}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791536343.8532128}
{"ep": "h2-00850", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.1709, "speedup": 1.1709, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 462.5, "calls": 2, "tokens_in": 30997, "tokens_out": 19083, "ctx_chars": 2425, "t_end": 1791536498.881726, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.1709, "lo95": 1.1684, "mean": 1.1695, "ev_only": 1.1727, "ref_ms": 18.86295, "cand_ms": 16.10974, "n_trials": 77, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1416, "r_wall_s": 23.08, "compile_s": 2.14, "r_error": "", "src_len": 16742, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 11747, "ts": 1791536127.5201159, "chk": "ok", "code_len": 16828}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19447, "tout": 7336, "ts": 1791536475.7434738, "chk": null, "code_len": 16742}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791536036.418249}
{"ep": "h2-00985", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0062, "speedup": 1.0062, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 48.0, "calls": 2, "tokens_in": 13713, "tokens_out": 1160, "ctx_chars": 10909, "t_end": 1791536502.0112834, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0062, "lo95": 1.0056, "mean": 1.0334, "ev_only": 1.0066, "ref_ms": 8.05303, "cand_ms": 8.00328, "n_trials": 146, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0038, "r_wall_s": 21.07, "compile_s": 2.13, "r_error": "", "src_len": 916, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6479, "tout": 646, "ts": 1791536468.0434093, "chk": "ok", "code_len": 916}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7234, "tout": 514, "ts": 1791536480.870059, "chk": null, "code_len": 916}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6479, "started": 1791536453.9829292}
{"ep": "h2-00980", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0055, "speedup": 1.0055, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 76.9, "calls": 2, "tokens_in": 4024, "tokens_out": 2138, "ctx_chars": 0, "t_end": 1791536504.0605974, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0055, "lo95": 1.0053, "mean": 1.0053, "ev_only": 1.0062, "ref_ms": 7.97947, "cand_ms": 7.93553, "n_trials": 160, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0043, "r_wall_s": 30.93, "compile_s": 1.94, "r_error": "", "src_len": 942, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1516, "ts": 1791536441.8605094, "chk": "ok", "code_len": 942}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2521, "tout": 622, "ts": 1791536473.0304945, "chk": null, "code_len": 942}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791536427.129693}
{"ep": "h2-00973", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0662, "speedup": 1.0662, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 102.5, "calls": 3, "tokens_in": 16423, "tokens_out": 2313, "ctx_chars": 6513, "t_end": 1791536511.1724458, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0662, "lo95": 1.0658, "mean": 1.0719, "ev_only": 1.0664, "ref_ms": 15.47374, "cand_ms": 14.51324, "n_trials": 84, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1175, "r_wall_s": 62.25, "compile_s": 2.63, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1158, "ts": 1791536418.7322142, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5875, "tout": 44, "ts": 1791536439.0808578, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5943, "tout": 1111, "ts": 1791536448.1976664, "chk": null, "code_len": 2422}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4605, "started": 1791536408.6302385}
{"ep": "h2-00934", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0046, "speedup": 1.0046, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 197.5, "calls": 2, "tokens_in": 4065, "tokens_out": 2115, "ctx_chars": 0, "t_end": 1791536511.6867976, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0046, "lo95": 1.0035, "mean": 1.0072, "ev_only": 1.0073, "ref_ms": 8.00875, "cand_ms": 7.97183, "n_trials": 151, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 131.97, "compile_s": 2.39, "r_error": "", "src_len": 993, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1503, "tout": 1461, "ts": 1791536336.334453, "chk": "ok", "code_len": 1028}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2562, "tout": 654, "ts": 1791536379.458662, "chk": null, "code_len": 993}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1503, "started": 1791536314.189592}
{"ep": "h2-00918", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 10.0, "speedup": 24.8376, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 261.0, "calls": 2, "tokens_in": 16304, "tokens_out": 8752, "ctx_chars": 10527, "t_end": 1791536511.7062762, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 24.8376, "lo95": 22.9376, "mean": 24.0651, "ev_only": 24.7835, "ref_ms": 19.77706, "cand_ms": 0.79626, "n_trials": 80, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "triton", "own_frac": 1.0, "r_wall_s": 137.32, "compile_s": 2.94, "r_error": "", "src_len": 5358, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 6345, "ts": 1791536322.9710977, "chk": "ok", "code_len": 5358}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9641, "tout": 2407, "ts": 1791536374.342143, "chk": null, "code_len": 5358}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791536250.6867795}
{"ep": "h2-00987", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 57.4, "calls": 1, "tokens_in": 1666, "tokens_out": 658, "ctx_chars": 0, "t_end": 1791536514.4202754, "error": "RuntimeError: Error building extension 'mt_fuse': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mt_fuse -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval-pch", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 50.33, "compile_s": 50.32, "r_error": "RuntimeError: Error building extension 'mt_fuse': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=mt_fuse -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval-pch", "src_len": 1324, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1666, "tout": 658, "ts": 1791536464.0692642, "chk": null, "code_len": 1324}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1666, "started": 1791536457.0595827}
{"ep": "h2-00857", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8358, "speedup": 0.8358, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 453.9, "calls": 4, "tokens_in": 83187, "tokens_out": 21062, "ctx_chars": 2425, "t_end": 1791536515.4705224, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.8358, "lo95": 0.8322, "mean": 0.8346, "ev_only": 0.836, "ref_ms": 18.94766, "cand_ms": 22.67133, "n_trials": 53, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0227, "r_wall_s": 18.44, "compile_s": 2.61, "r_error": "", "src_len": 16602, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7119, "ts": 1791536118.1298544, "chk": "runtime_error", "code_len": 16649}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 19130, "tout": 6978, "ts": 1791536432.1669884, "chk": "ok", "code_len": 16602}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 26220, "tout": 45, "ts": 1791536442.7345366, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 26289, "tout": 6920, "ts": 1791536496.9988241, "chk": null, "code_len": 16602}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["runtime_error", "ok", "no_code", null], "first_tin": 11548, "started": 1791536061.5656636}
{"ep": "h2-00975", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9935, "speedup": 0.9935, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 105.5, "calls": 3, "tokens_in": 22602, "tokens_out": 3066, "ctx_chars": 10823, "t_end": 1791536517.766346, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9935, "lo95": 0.9926, "mean": 0.9933, "ev_only": 0.993, "ref_ms": 30.109, "cand_ms": 30.30543, "n_trials": 52, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0186, "r_wall_s": 62.83, "compile_s": 2.31, "r_error": "", "src_len": 2884, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1621, "ts": 1791536426.8163712, "chk": "ok", "code_len": 2884}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8086, "tout": 53, "ts": 1791536442.7542675, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8163, "tout": 1392, "ts": 1791536454.4909737, "chk": null, "code_len": 2884}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791536412.2235484}
{"ep": "h2-00971", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0276, "speedup": 1.0276, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 113.6, "calls": 2, "tokens_in": 4480, "tokens_out": 2303, "ctx_chars": 0, "t_end": 1791536520.7061307, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0276, "lo95": 1.026, "mean": 1.0481, "ev_only": 1.0279, "ref_ms": 24.86574, "cand_ms": 24.19819, "n_trials": 55, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0227, "r_wall_s": 76.5, "compile_s": 1.96, "r_error": "", "src_len": 1054, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 1631, "ts": 1791536421.5904992, "chk": "ok", "code_len": 1068}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2812, "tout": 672, "ts": 1791536443.4806168, "chk": null, "code_len": 1054}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791536407.093163}
{"ep": "h2-00998", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "incorrect", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 40.4, "calls": 1, "tokens_in": 1633, "tokens_out": 3160, "ctx_chars": 0, "t_end": 1791536522.700005, "error": "output shape (1024, 8192, 1, 1) != expected (1, 8192, 1024, 1)", "f_status": "incorrect", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "triton", "own_frac": 0.004, "r_wall_s": 11.2, "compile_s": 2.36, "r_error": "output shape (1024, 8192, 1, 1) != expected (1, 8192, 1024, 1)", "src_len": 1541, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1633, "tout": 3160, "ts": 1791536511.4445286, "chk": null, "code_len": 1541}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1633, "started": 1791536482.2928786}
{"ep": "h2-00957", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.028, "speedup": 1.028, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 159.3, "calls": 2, "tokens_in": 13798, "tokens_out": 11713, "ctx_chars": 6513, "t_end": 1791536526.398282, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.028, "lo95": 1.0277, "mean": 1.028, "ev_only": 1.0272, "ref_ms": 15.69315, "cand_ms": 15.26557, "n_trials": 82, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.5636, "r_wall_s": 36.57, "compile_s": 2.45, "r_error": "", "src_len": 7092, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 8270, "ts": 1791536438.8786974, "chk": "ok", "code_len": 7092}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8934, "tout": 3443, "ts": 1791536489.4766493, "chk": null, "code_len": 7092}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791536367.0751705}
{"ep": "h2-00991", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0189, "speedup": 1.0189, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 59.8, "calls": 1, "tokens_in": 1750, "tokens_out": 1879, "ctx_chars": 0, "t_end": 1791536526.6456625, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0189, "lo95": 1.0182, "mean": 1.0179, "ev_only": 1.0195, "ref_ms": 8.14762, "cand_ms": 7.99636, "n_trials": 154, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0089, "r_wall_s": 37.86, "compile_s": 2.65, "r_error": "", "src_len": 1300, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1750, "tout": 1879, "ts": 1791536488.7255373, "chk": null, "code_len": 1300}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1750, "started": 1791536466.8698866}
{"ep": "h2-00944", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.0238, "speedup": 2.0238, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 192.7, "calls": 2, "tokens_in": 7818, "tokens_out": 9323, "ctx_chars": 0, "t_end": 1791536530.8537288, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 2.0238, "lo95": 2.0191, "mean": 2.0149, "ev_only": 2.0199, "ref_ms": 14.36445, "cand_ms": 7.09791, "n_trials": 110, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 46.81, "compile_s": 15.04, "r_error": "", "src_len": 5857, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6450, "ts": 1791536432.697736, "chk": "ok", "code_len": 5857}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5635, "tout": 2873, "ts": 1791536483.9490092, "chk": null, "code_len": 5857}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2183, "started": 1791536338.1997185}
{"ep": "h2-00923", "task": "L3_014", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0633, "speedup": 1.0633, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 262.5, "calls": 2, "tokens_in": 10519, "tokens_out": 2320, "ctx_chars": 6513, "t_end": 1791536532.659196, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0633, "lo95": 1.0629, "mean": 1.0628, "ev_only": 1.0638, "ref_ms": 15.47229, "cand_ms": 14.55147, "n_trials": 109, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0683, "r_wall_s": 143.01, "compile_s": 2.08, "r_error": "", "src_len": 2422, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4605, "tout": 1197, "ts": 1791536281.3530438, "chk": "ok", "code_len": 2422}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5914, "tout": 1123, "ts": 1791536388.9478087, "chk": null, "code_len": 2422}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4605, "started": 1791536270.1430452}
{"ep": "h2-00978", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9218, "speedup": 0.9218, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 114.4, "calls": 2, "tokens_in": 15372, "tokens_out": 1891, "ctx_chars": 11539, "t_end": 1791536532.989535, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9218, "lo95": 0.9215, "mean": 0.9218, "ev_only": 0.9223, "ref_ms": 7.75634, "cand_ms": 8.41399, "n_trials": 199, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.077, "r_wall_s": 55.47, "compile_s": 1.88, "r_error": "", "src_len": 1736, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1037, "ts": 1791536435.8755496, "chk": "ok", "code_len": 1736}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8259, "tout": 854, "ts": 1791536476.7942576, "chk": null, "code_len": 1736}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7113, "started": 1791536419.1730783}
{"ep": "h2-00933", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7771, "speedup": 0.7771, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 232.3, "calls": 3, "tokens_in": 23488, "tokens_out": 2801, "ctx_chars": 11065, "t_end": 1791536544.9531527, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 0.7771, "lo95": 0.7753, "mean": 0.8022, "ev_only": 0.777, "ref_ms": 1.81346, "cand_ms": 2.33355, "n_trials": 207, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.4631, "r_wall_s": 153.99, "compile_s": 2.06, "r_error": "", "src_len": 2284, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 1665, "ts": 1791536332.8152783, "chk": "ok", "code_len": 2427}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8404, "tout": 29, "ts": 1791536381.5559998, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8457, "tout": 1107, "ts": 1791536390.9403763, "chk": null, "code_len": 2284}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6627, "started": 1791536312.6378157}
{"ep": "h2-00936", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.3614, "speedup": 2.3614, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 230.9, "calls": 3, "tokens_in": 20927, "tokens_out": 3895, "ctx_chars": 8529, "t_end": 1791536549.4025679, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 2.3614, "lo95": 2.3604, "mean": 2.3597, "ev_only": 2.3726, "ref_ms": 8.47971, "cand_ms": 3.59102, "n_trials": 193, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4104, "r_wall_s": 159.35, "compile_s": 2.05, "r_error": "", "src_len": 3472, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 2084, "ts": 1791536358.5181696, "chk": "ok", "code_len": 3472}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7691, "tout": 26, "ts": 1791536375.5270355, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7741, "tout": 1785, "ts": 1791536390.0070436, "chk": null, "code_len": 3472}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5495, "started": 1791536318.4987476}
{"ep": "h2-00917", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 6.6906, "speedup": 6.6906, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 310.5, "calls": 3, "tokens_in": 18154, "tokens_out": 19177, "ctx_chars": 0, "t_end": 1791536558.6936967, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 6.6906, "lo95": 6.5582, "mean": 7.9494, "ev_only": 6.7486, "ref_ms": 33.54263, "cand_ms": 5.01342, "n_trials": 33, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9911, "r_wall_s": 67.28, "compile_s": 12.59, "r_error": "", "src_len": 7494, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 11452, "ts": 1791536360.1318316, "chk": "no_custom_kernel", "code_len": 7519}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6193, "tout": 4114, "ts": 1791536450.2531033, "chk": "ok", "code_len": 7494}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9995, "tout": 3611, "ts": 1791536491.15806, "chk": null, "code_len": 7494}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791536248.1454422}
{"ep": "h2-00950", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 0.7574, "speedup": 0.7574, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 218.2, "calls": 2, "tokens_in": 12049, "tokens_out": 18637, "ctx_chars": 0, "t_end": 1791536570.6475563, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7574, "lo95": 0.7091, "mean": 0.8725, "ev_only": 0.7578, "ref_ms": 7.83924, "cand_ms": 10.35025, "n_trials": 133, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.994, "r_wall_s": 46.27, "compile_s": 2.36, "r_error": "", "src_len": 11129, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 13100, "ts": 1791536463.9158676, "chk": "ok", "code_len": 11129}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9105, "tout": 5537, "ts": 1791536524.3533003, "chk": null, "code_len": 11129}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791536352.427998}
{"ep": "h2-01011", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0518, "speedup": 1.0518, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 46.5, "calls": 1, "tokens_in": 4862, "tokens_out": 1878, "ctx_chars": 6513, "t_end": 1791536572.8696196, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0518, "lo95": 1.0516, "mean": 1.0518, "ev_only": 1.0522, "ref_ms": 15.69084, "cand_ms": 14.91806, "n_trials": 95, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1722, "r_wall_s": 30.98, "compile_s": 2.95, "r_error": "", "src_len": 3775, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 4862, "tout": 1878, "ts": 1791536541.6671538, "chk": null, "code_len": 3775}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 4862, "started": 1791536526.39854}
{"ep": "h2-00988", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.922, "speedup": 0.922, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 124.7, "calls": 3, "tokens_in": 24184, "tokens_out": 2235, "ctx_chars": 11539, "t_end": 1791536584.4633646, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.922, "lo95": 0.9214, "mean": 0.9218, "ev_only": 0.9228, "ref_ms": 7.75887, "cand_ms": 8.4157, "n_trials": 136, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0604, "r_wall_s": 75.92, "compile_s": 2.0, "r_error": "", "src_len": 1919, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1266, "ts": 1791536481.1465755, "chk": "ok", "code_len": 1919}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8488, "tout": 71, "ts": 1791536499.7195017, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8583, "tout": 898, "ts": 1791536507.4696333, "chk": null, "code_len": 1919}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791536459.8102388}
{"ep": "h2-00972", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3818, "speedup": 1.3818, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 180.2, "calls": 2, "tokens_in": 16679, "tokens_out": 7882, "ctx_chars": 11539, "t_end": 1791536587.3873727, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.3818, "lo95": 1.3814, "mean": 1.3818, "ev_only": 1.3824, "ref_ms": 31.48763, "cand_ms": 22.7876, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9602, "r_wall_s": 90.64, "compile_s": 2.02, "r_error": "", "src_len": 4251, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 5695, "ts": 1791536456.429348, "chk": "ok", "code_len": 4251}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9645, "tout": 2187, "ts": 1791536495.887253, "chk": null, "code_len": 4251}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791536407.2374198}
{"ep": "h2-01005", "task": "L2_029", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0017, "speedup": 1.0017, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 79.3, "calls": 2, "tokens_in": 11938, "tokens_out": 1730, "ctx_chars": 9028, "t_end": 1791536591.054482, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0017, "lo95": 1.0014, "mean": 1.0507, "ev_only": 1.003, "ref_ms": 7.98339, "cand_ms": 7.96997, "n_trials": 121, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0075, "r_wall_s": 17.52, "compile_s": 1.86, "r_error": "", "src_len": 1076, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5371, "tout": 1084, "ts": 1791536525.6709888, "chk": "ok", "code_len": 1076}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6567, "tout": 646, "ts": 1791536573.443726, "chk": null, "code_len": 1076}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5371, "started": 1791536511.706479}
{"ep": "h2-00937", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8739, "speedup": 0.8739, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 273.2, "calls": 3, "tokens_in": 46311, "tokens_out": 13999, "ctx_chars": 0, "t_end": 1791536592.1856794, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.8739, "lo95": 0.5683, "mean": 0.9136, "ev_only": 0.8734, "ref_ms": 22.49414, "cand_ms": 25.73948, "n_trials": 24, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0268, "r_wall_s": 66.03, "compile_s": 2.85, "r_error": "", "src_len": 16593, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10632, "tout": 7077, "ts": 1791536376.8831959, "chk": "ok", "code_len": 16620}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 17821, "tout": 13, "ts": 1791536471.9505053, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 17858, "tout": 6909, "ts": 1791536526.0962577, "chk": null, "code_len": 16593}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 10632, "started": 1791536319.0316815}
{"ep": "h2-00959", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4453, "speedup": 1.4453, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 222.2, "calls": 2, "tokens_in": 11188, "tokens_out": 5053, "ctx_chars": 6513, "t_end": 1791536595.1454837, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.4453, "lo95": 1.4439, "mean": 1.4323, "ev_only": 1.4466, "ref_ms": 15.57291, "cand_ms": 10.77484, "n_trials": 96, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1908, "r_wall_s": 97.54, "compile_s": 2.04, "r_error": "", "src_len": 3468, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3498, "ts": 1791536413.6086774, "chk": "ok", "code_len": 3468}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6581, "tout": 1555, "ts": 1791536496.622243, "chk": null, "code_len": 3468}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791536372.902896}
{"ep": "h2-00889", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0009, "speedup": 1.0009, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 448.6, "calls": 2, "tokens_in": 4235, "tokens_out": 1981, "ctx_chars": 0, "t_end": 1791536595.3961794, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0009, "lo95": 1.0003, "mean": 1.0011, "ev_only": 0.9996, "ref_ms": 7.96617, "cand_ms": 7.95931, "n_trials": 196, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 194.77, "compile_s": 2.28, "r_error": "", "src_len": 862, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 1465, "ts": 1791536160.6051602, "chk": "ok", "code_len": 862}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2611, "tout": 516, "ts": 1791536400.5559225, "chk": null, "code_len": 862}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791536146.7790213}
{"ep": "h2-00976", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3683, "speedup": 1.3683, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 180.2, "calls": 2, "tokens_in": 17501, "tokens_out": 8265, "ctx_chars": 11539, "t_end": 1791536596.2545054, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.3683, "lo95": 1.3673, "mean": 1.368, "ev_only": 1.3698, "ref_ms": 7.77689, "cand_ms": 5.68347, "n_trials": 201, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 96.1, "compile_s": 2.02, "r_error": "", "src_len": 5303, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 5552, "ts": 1791536462.47101, "chk": "ok", "code_len": 5303}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10386, "tout": 2713, "ts": 1791536499.2393212, "chk": null, "code_len": 5303}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7115, "started": 1791536416.0895858}
{"ep": "h2-00986", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0485, "speedup": 1.0485, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 141.4, "calls": 3, "tokens_in": 18846, "tokens_out": 3947, "ctx_chars": 6513, "t_end": 1791536598.2059743, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0485, "lo95": 1.0479, "mean": 1.0688, "ev_only": 1.05, "ref_ms": 15.78145, "cand_ms": 15.0511, "n_trials": 85, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.144, "r_wall_s": 72.54, "compile_s": 2.74, "r_error": "", "src_len": 3966, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1978, "ts": 1791536481.512109, "chk": "ok", "code_len": 3966}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6952, "tout": 56, "ts": 1791536509.0599017, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7032, "tout": 1913, "ts": 1791536525.4740565, "chk": null, "code_len": 3966}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791536456.7695878}
{"ep": "h2-00994", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0065, "speedup": 1.0065, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 135.7, "calls": 3, "tokens_in": 25023, "tokens_out": 4526, "ctx_chars": 10167, "t_end": 1791536604.8514318, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0065, "lo95": 1.0059, "mean": 1.0082, "ev_only": 1.0072, "ref_ms": 8.44874, "cand_ms": 8.3939, "n_trials": 140, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0563, "r_wall_s": 20.62, "compile_s": 1.89, "r_error": "", "src_len": 2564, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3210, "ts": 1791536497.5813153, "chk": "ok", "code_len": 2659}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9426, "tout": 43, "ts": 1791536573.3634036, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9493, "tout": 1273, "ts": 1791536584.136659, "chk": null, "code_len": 2564}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791536469.1850617}
{"ep": "h2-00876", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0494, "speedup": 1.0494, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 498.9, "calls": 2, "tokens_in": 4915, "tokens_out": 3558, "ctx_chars": 0, "t_end": 1791536607.380071, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0494, "lo95": 1.0477, "mean": 1.0251, "ev_only": 1.0546, "ref_ms": 8.50483, "cand_ms": 8.10463, "n_trials": 68, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.006, "r_wall_s": 215.76, "compile_s": 2.0, "r_error": "", "src_len": 2329, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 2246, "ts": 1791536127.49911, "chk": "ok", "code_len": 2329}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3280, "tout": 1312, "ts": 1791536391.5384517, "chk": null, "code_len": 2329}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791536108.5120475}
{"ep": "h2-01017", "task": "L2_076", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0039, "speedup": 1.0039, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 61.9, "calls": 1, "tokens_in": 6367, "tokens_out": 541, "ctx_chars": 10847, "t_end": 1791536611.292141, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0039, "lo95": 1.0015, "mean": 1.0043, "ev_only": 1.0053, "ref_ms": 7.98598, "cand_ms": 7.95481, "n_trials": 153, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0039, "r_wall_s": 47.43, "compile_s": 1.83, "r_error": "", "src_len": 1003, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6367, "tout": 541, "ts": 1791536563.7032557, "chk": null, "code_len": 1003}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6367, "started": 1791536549.4028397}
{"ep": "h2-00842", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 7.6704, "speedup": 7.6704, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 601.6, "calls": 3, "tokens_in": 19259, "tokens_out": 24381, "ctx_chars": 0, "t_end": 1791536611.6016695, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 7.6704, "lo95": 5.6194, "mean": 7.0391, "ev_only": 7.7099, "ref_ms": 45.81642, "cand_ms": 5.97313, "n_trials": 37, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.987, "r_wall_s": 64.96, "compile_s": 11.76, "r_error": "", "src_len": 8477, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 15919, "ts": 1791536160.325896, "chk": "no_custom_kernel", "code_len": 8519}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6515, "tout": 4488, "ts": 1791536369.9775865, "chk": "ok", "code_len": 8477}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10778, "tout": 3974, "ts": 1791536546.5950406, "chk": null, "code_len": 8477}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791536009.9596293}
{"ep": "h2-00982", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 3.9575, "speedup": 3.9575, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 175.4, "calls": 2, "tokens_in": 6108, "tokens_out": 6715, "ctx_chars": 0, "t_end": 1791536614.0613604, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 3.9575, "lo95": 3.8994, "mean": 3.8, "ev_only": 4.0819, "ref_ms": 1.85238, "cand_ms": 0.46807, "n_trials": 201, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 99.32, "compile_s": 14.96, "r_error": "", "src_len": 4188, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4587, "ts": 1791536477.640908, "chk": "ok", "code_len": 4188}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4349, "tout": 2128, "ts": 1791536514.4853363, "chk": null, "code_len": 4188}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1759, "started": 1791536438.650034}
{"ep": "h2-00990", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0618, "speedup": 1.0618, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 150.6, "calls": 2, "tokens_in": 16032, "tokens_out": 4778, "ctx_chars": 10167, "t_end": 1791536616.6763966, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0618, "lo95": 1.061, "mean": 1.062, "ev_only": 1.0595, "ref_ms": 8.48291, "cand_ms": 7.98925, "n_trials": 139, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0057, "r_wall_s": 93.01, "compile_s": 2.66, "r_error": "", "src_len": 2118, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3712, "ts": 1791536501.2530246, "chk": "ok", "code_len": 2118}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9928, "tout": 1066, "ts": 1791536523.360986, "chk": null, "code_len": 2118}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6104, "started": 1791536466.1129465}
{"ep": "h2-00940", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0038, "speedup": 1.0038, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 303.3, "calls": 3, "tokens_in": 20299, "tokens_out": 2872, "ctx_chars": 9266, "t_end": 1791536625.9832902, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0038, "lo95": 1.0016, "mean": 0.9999, "ev_only": 1.0057, "ref_ms": 30.5079, "cand_ms": 30.39187, "n_trials": 24, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0399, "r_wall_s": 262.93, "compile_s": 2.85, "r_error": "", "src_len": 2500, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1577, "ts": 1791536337.6074202, "chk": "ok", "code_len": 2500}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7303, "tout": 55, "ts": 1791536351.890494, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7382, "tout": 1240, "ts": 1791536362.3932076, "chk": null, "code_len": 2500}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791536322.667852}
{"ep": "h2-01014", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0914, "speedup": 1.0914, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 95.5, "calls": 2, "tokens_in": 15457, "tokens_out": 4367, "ctx_chars": 10527, "t_end": 1791536628.136523, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0914, "lo95": 1.0905, "mean": 1.1174, "ev_only": 1.0919, "ref_ms": 19.63914, "cand_ms": 17.99451, "n_trials": 88, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0397, "r_wall_s": 54.07, "compile_s": 1.89, "r_error": "", "src_len": 3227, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6663, "tout": 2728, "ts": 1791536555.0784185, "chk": "ok", "code_len": 3227}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8794, "tout": 1639, "ts": 1791536573.9362817, "chk": null, "code_len": 3227}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6663, "started": 1791536532.659452}
{"ep": "h2-00953", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9941, "speedup": 0.9941, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 270.9, "calls": 3, "tokens_in": 22403, "tokens_out": 2967, "ctx_chars": 10823, "t_end": 1791536628.6769109, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.9941, "lo95": 0.9938, "mean": 0.994, "ev_only": 0.9941, "ref_ms": 30.09545, "cand_ms": 30.27477, "n_trials": 38, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0386, "r_wall_s": 87.35, "compile_s": 1.96, "r_error": "", "src_len": 2872, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1521, "ts": 1791536371.7073143, "chk": "ok", "code_len": 2872}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7986, "tout": 54, "ts": 1791536529.3763354, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8064, "tout": 1392, "ts": 1791536540.8518655, "chk": null, "code_len": 2872}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791536358.4846773}
{"ep": "h2-00927", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4253, "speedup": 1.4253, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 344.0, "calls": 2, "tokens_in": 16864, "tokens_out": 7999, "ctx_chars": 11539, "t_end": 1791536629.4101443, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.4253, "lo95": 1.425, "mean": 1.4252, "ev_only": 1.4257, "ref_ms": 31.40833, "cand_ms": 22.0364, "n_trials": 48, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9632, "r_wall_s": 234.66, "compile_s": 2.05, "r_error": "", "src_len": 4021, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7034, "tout": 5788, "ts": 1791536335.4956675, "chk": "ok", "code_len": 4021}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9830, "tout": 2211, "ts": 1791536393.97078, "chk": null, "code_len": 4021}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7034, "started": 1791536285.3933754}
{"ep": "h2-00961", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.7823, "speedup": 0.7823, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.1, "calls": 3, "tokens_in": 49842, "tokens_out": 14679, "ctx_chars": 2425, "t_end": 1791536630.490414, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.7823, "lo95": 0.58, "mean": 0.7692, "ev_only": 0.7817, "ref_ms": 26.01039, "cand_ms": 33.24771, "n_trials": 28, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0297, "r_wall_s": 102.71, "compile_s": 3.01, "r_error": "", "src_len": 17147, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7439, "ts": 1791536439.1716573, "chk": "ok", "code_len": 17147}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19099, "tout": 72, "ts": 1791536454.6719675, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19195, "tout": 7168, "ts": 1791536527.4103153, "chk": null, "code_len": 17147}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791536378.3484342}
{"ep": "h2-00952", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 3.3057, "speedup": 3.3057, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 274.0, "calls": 3, "tokens_in": 15884, "tokens_out": 17555, "ctx_chars": 0, "t_end": 1791536631.1134963, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 3.3057, "lo95": 3.2842, "mean": 3.7727, "ev_only": 3.3116, "ref_ms": 60.73701, "cand_ms": 18.37315, "n_trials": 27, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8658, "r_wall_s": 44.87, "compile_s": 10.81, "r_error": "", "src_len": 6045, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 11245, "ts": 1791536459.458106, "chk": "no_custom_kernel", "code_len": 6048}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5379, "tout": 3299, "ts": 1791536497.8204997, "chk": "ok", "code_len": 6045}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8519, "tout": 3011, "ts": 1791536586.1057813, "chk": null, "code_len": 6045}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791536357.1514618}
{"ep": "h2-00993", "task": "L2_028", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0163, "speedup": 1.0163, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 164.2, "calls": 2, "tokens_in": 11393, "tokens_out": 4383, "ctx_chars": 5026, "t_end": 1791536631.6789975, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0163, "lo95": 1.0145, "mean": 1.0128, "ev_only": 1.0194, "ref_ms": 8.16069, "cand_ms": 8.03001, "n_trials": 98, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0051, "r_wall_s": 115.87, "compile_s": 2.99, "r_error": "", "src_len": 2076, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3993, "tout": 3295, "ts": 1791536495.874051, "chk": "ok", "code_len": 2076}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7400, "tout": 1088, "ts": 1791536515.5043283, "chk": null, "code_len": 2076}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3993, "started": 1791536467.4843209}
{"ep": "h2-01000", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0111, "speedup": 1.0111, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 134.4, "calls": 2, "tokens_in": 16179, "tokens_out": 2495, "ctx_chars": 12117, "t_end": 1791536633.2553666, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0111, "lo95": 1.0104, "mean": 1.011, "ev_only": 1.0111, "ref_ms": 42.50706, "cand_ms": 42.04064, "n_trials": 40, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 62.16, "compile_s": 1.88, "r_error": "", "src_len": 1956, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1542, "ts": 1791536511.9496071, "chk": "ok", "code_len": 1956}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8780, "tout": 953, "ts": 1791536570.3628688, "chk": null, "code_len": 1956}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791536498.8819919}
{"ep": "h2-01007", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9199, "speedup": 0.9199, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 121.0, "calls": 3, "tokens_in": 23744, "tokens_out": 1900, "ctx_chars": 11539, "t_end": 1791536636.4695349, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9199, "lo95": 0.9185, "mean": 0.9203, "ev_only": 0.9208, "ref_ms": 7.82168, "cand_ms": 8.50288, "n_trials": 154, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0609, "r_wall_s": 67.57, "compile_s": 2.13, "r_error": "", "src_len": 1777, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1075, "ts": 1791536526.9185026, "chk": "ok", "code_len": 1777}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8297, "tout": 13, "ts": 1791536559.885174, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8334, "tout": 812, "ts": 1791536568.1058939, "chk": null, "code_len": 1777}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791536515.470787}
{"ep": "h2-00992", "task": "L2_047", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.057, "speedup": 1.057, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 170.8, "calls": 2, "tokens_in": 4721, "tokens_out": 2977, "ctx_chars": 0, "t_end": 1791536637.888007, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.057, "lo95": 1.0556, "mean": 1.0567, "ev_only": 1.0571, "ref_ms": 24.87142, "cand_ms": 23.53087, "n_trials": 64, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0278, "r_wall_s": 110.08, "compile_s": 3.12, "r_error": "", "src_len": 1398, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1668, "tout": 2153, "ts": 1791536501.2316916, "chk": "ok", "code_len": 1398}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3053, "tout": 824, "ts": 1791536526.8301194, "chk": null, "code_len": 1398}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1668, "started": 1791536467.1181035}
{"ep": "h2-00942", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.2325, "speedup": 2.2325, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 312.9, "calls": 2, "tokens_in": 6674, "tokens_out": 8215, "ctx_chars": 0, "t_end": 1791536638.3226576, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.2325, "lo95": 2.2299, "mean": 2.2327, "ev_only": 2.2349, "ref_ms": 42.6477, "cand_ms": 19.10349, "n_trials": 33, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9998, "r_wall_s": 75.91, "compile_s": 2.05, "r_error": "", "src_len": 4154, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 5676, "ts": 1791536377.68515, "chk": "ok", "code_len": 4154}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4638, "tout": 2539, "ts": 1791536561.739108, "chk": null, "code_len": 4154}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2036, "started": 1791536325.4601228}
{"ep": "h2-00958", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.1236, "speedup": 2.1236, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 267.8, "calls": 3, "tokens_in": 16122, "tokens_out": 14761, "ctx_chars": 0, "t_end": 1791536640.2450917, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 2.1236, "lo95": 2.1225, "mean": 2.1258, "ev_only": 2.1242, "ref_ms": 59.71654, "cand_ms": 28.12006, "n_trials": 24, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9115, "r_wall_s": 45.19, "compile_s": 12.69, "r_error": "", "src_len": 6029, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 8294, "ts": 1791536454.6932297, "chk": "no_custom_kernel", "code_len": 6030}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5476, "tout": 3498, "ts": 1791536529.6171663, "chk": "ok", "code_len": 6029}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8660, "tout": 2969, "ts": 1791536594.9358425, "chk": null, "code_len": 6029}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791536372.4364922}
{"ep": "h2-00945", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.196, "speedup": 1.196, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 301.9, "calls": 2, "tokens_in": 16693, "tokens_out": 17751, "ctx_chars": 6513, "t_end": 1791536643.814276, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.196, "lo95": 1.1803, "mean": 1.2793, "ev_only": 1.2001, "ref_ms": 7.05649, "cand_ms": 5.90022, "n_trials": 129, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.4819, "r_wall_s": 47.44, "compile_s": 11.5, "r_error": "", "src_len": 11164, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5647, "tout": 12813, "ts": 1791536453.2944076, "chk": "ok", "code_len": 11246}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 11046, "tout": 4938, "ts": 1791536596.3295186, "chk": null, "code_len": 11164}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5647, "started": 1791536341.9186277}
{"ep": "h2-00981", "task": "L1_056", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.5176, "speedup": 1.5176, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 213.1, "calls": 2, "tokens_in": 6753, "tokens_out": 7121, "ctx_chars": 0, "t_end": 1791536645.146297, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.5176, "lo95": 1.5168, "mean": 1.5218, "ev_only": 1.5186, "ref_ms": 31.3926, "cand_ms": 20.68552, "n_trials": 37, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9997, "r_wall_s": 128.21, "compile_s": 1.92, "r_error": "", "src_len": 3726, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2078, "tout": 5086, "ts": 1791536480.4179046, "chk": "ok", "code_len": 3726}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4675, "tout": 2035, "ts": 1791536515.492734, "chk": null, "code_len": 3726}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2078, "started": 1791536432.061288}
{"ep": "h2-01004", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0034, "speedup": 1.0034, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 133.7, "calls": 3, "tokens_in": 24738, "tokens_out": 2201, "ctx_chars": 12117, "t_end": 1791536645.3449893, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0034, "lo95": 1.0021, "mean": 1.0036, "ev_only": 1.0036, "ref_ms": 42.58493, "cand_ms": 42.44156, "n_trials": 38, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 79.15, "compile_s": 2.0, "r_error": "", "src_len": 2279, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1126, "ts": 1791536523.9231498, "chk": "ok", "code_len": 2279}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8635, "tout": 47, "ts": 1791536556.9670434, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8706, "tout": 1028, "ts": 1791536565.5283267, "chk": null, "code_len": 2279}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791536511.6870353}
{"ep": "h2-01008", "task": "L2_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0039, "speedup": 1.0039, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 144.5, "calls": 3, "tokens_in": 20338, "tokens_out": 2983, "ctx_chars": 9266, "t_end": 1791536662.2248077, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0039, "lo95": 1.003, "mean": 1.0042, "ev_only": 1.0045, "ref_ms": 30.40391, "cand_ms": 30.28683, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0425, "r_wall_s": 63.64, "compile_s": 2.12, "r_error": "", "src_len": 2630, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5614, "tout": 1598, "ts": 1791536534.1442568, "chk": "ok", "code_len": 2630}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7324, "tout": 52, "ts": 1791536587.1332211, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7400, "tout": 1333, "ts": 1791536597.9946797, "chk": null, "code_len": 2630}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5614, "started": 1791536517.7665966}
{"ep": "h2-01015", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.1996, "speedup": 1.1996, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 134.1, "calls": 2, "tokens_in": 5107, "tokens_out": 3957, "ctx_chars": 0, "t_end": 1791536667.1115582, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.1996, "lo95": 1.198, "mean": 1.2011, "ev_only": 1.2001, "ref_ms": 8.05591, "cand_ms": 6.71533, "n_trials": 174, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 41.38, "compile_s": 3.78, "r_error": "", "src_len": 2611, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1624, "tout": 2533, "ts": 1791536555.1458225, "chk": "ok", "code_len": 2611}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3483, "tout": 1424, "ts": 1791536625.612445, "chk": null, "code_len": 2611}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1624, "started": 1791536532.9898164}
{"ep": "h2-00887", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3696, "speedup": 1.3696, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 523.1, "calls": 2, "tokens_in": 15858, "tokens_out": 20255, "ctx_chars": 6513, "t_end": 1791536669.1121998, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.3696, "lo95": 1.3683, "mean": 1.3711, "ev_only": 1.3696, "ref_ms": 15.70041, "cand_ms": 11.4638, "n_trials": 79, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.3743, "r_wall_s": 257.12, "compile_s": 57.72, "r_error": "", "src_len": 11353, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 14719, "ts": 1791536273.32068, "chk": "ok", "code_len": 11353}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 10994, "tout": 5536, "ts": 1791536411.447105, "chk": null, "code_len": 11353}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791536146.0547094}
{"ep": "h2-01031", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0005, "speedup": 1.0005, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.3, "calls": 1, "tokens_in": 6477, "tokens_out": 520, "ctx_chars": 10909, "t_end": 1791536688.5461974, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0005, "lo95": 0.9992, "mean": 1.0327, "ev_only": 1.0037, "ref_ms": 8.12805, "cand_ms": 8.12362, "n_trials": 167, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 71.88, "compile_s": 2.47, "r_error": "", "src_len": 976, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 6477, "tout": 520, "ts": 1791536616.5762157, "chk": null, "code_len": 976}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 6477, "started": 1791536611.292368}
{"ep": "h2-00951", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7878, "speedup": 0.7878, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 333.5, "calls": 3, "tokens_in": 50147, "tokens_out": 15047, "ctx_chars": 2425, "t_end": 1791536690.4649932, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.7878, "lo95": 0.7856, "mean": 0.8039, "ev_only": 0.7876, "ref_ms": 19.09932, "cand_ms": 24.24464, "n_trials": 61, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.03, "r_wall_s": 18.73, "compile_s": 1.96, "r_error": "", "src_len": 17661, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11548, "tout": 7610, "ts": 1791536417.3487175, "chk": "ok", "code_len": 17661}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19270, "tout": 35, "ts": 1791536614.3711703, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 19329, "tout": 7402, "ts": 1791536671.6701207, "chk": null, "code_len": 17661}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 11548, "started": 1791536357.0011053}
{"ep": "h2-01009", "task": "L3_021", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0497, "speedup": 1.0497, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 170.3, "calls": 3, "tokens_in": 18748, "tokens_out": 3859, "ctx_chars": 6513, "t_end": 1791536691.0448403, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0497, "lo95": 1.049, "mean": 1.0504, "ev_only": 1.0523, "ref_ms": 15.8768, "cand_ms": 15.12487, "n_trials": 68, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1443, "r_wall_s": 76.72, "compile_s": 2.37, "r_error": "", "src_len": 3925, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4862, "tout": 1944, "ts": 1791536536.62198, "chk": "ok", "code_len": 3925}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6918, "tout": 26, "ts": 1791536598.5512033, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6968, "tout": 1889, "ts": 1791536613.8926597, "chk": null, "code_len": 3925}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 4862, "started": 1791536520.706374}
{"ep": "h2-00964", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0268, "speedup": 1.0268, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 302.4, "calls": 1, "tokens_in": 5832, "tokens_out": 524, "ctx_chars": 9656, "t_end": 1791536691.3068764, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0268, "lo95": 1.0254, "mean": 1.0268, "ev_only": 1.0269, "ref_ms": 24.85462, "cand_ms": 24.20636, "n_trials": 65, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0223, "r_wall_s": 296.71, "compile_s": 2.04, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 524, "ts": 1791536393.8787487, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791536388.9022667}
{"ep": "h2-01006", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9931, "speedup": 0.9931, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 182.1, "calls": 3, "tokens_in": 22384, "tokens_out": 2950, "ctx_chars": 10823, "t_end": 1791536696.5115995, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9931, "lo95": 0.9921, "mean": 0.9935, "ev_only": 0.9946, "ref_ms": 30.12459, "cand_ms": 30.33534, "n_trials": 48, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0367, "r_wall_s": 70.65, "compile_s": 3.87, "r_error": "", "src_len": 2871, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1514, "ts": 1791536527.8265264, "chk": "ok", "code_len": 2871}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7979, "tout": 49, "ts": 1791536613.9832253, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8052, "tout": 1387, "ts": 1791536625.3262417, "chk": null, "code_len": 2871}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791536514.4205246}
{"ep": "h2-01032", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 86.3, "calls": 1, "tokens_in": 2157, "tokens_out": 1384, "ctx_chars": 0, "t_end": 1791536697.9044447, "error": "RuntimeError: Error building extension 'conv3d_bias_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=conv3d_bias_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 72.32, "compile_s": 72.32, "r_error": "RuntimeError: Error building extension 'conv3d_bias_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=conv3d_bias_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/", "src_len": 2670, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2157, "tout": 1384, "ts": 1791536625.5249631, "chk": null, "code_len": 2670}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2157, "started": 1791536611.6024325}
{"ep": "h2-01012", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9937, "speedup": 0.9937, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 178.9, "calls": 3, "tokens_in": 22378, "tokens_out": 2934, "ctx_chars": 10823, "t_end": 1791536705.574305, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9937, "lo95": 0.9923, "mean": 0.9934, "ev_only": 0.9937, "ref_ms": 30.07288, "cand_ms": 30.26349, "n_trials": 39, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0411, "r_wall_s": 132.95, "compile_s": 1.98, "r_error": "", "src_len": 2884, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1529, "ts": 1791536540.420118, "chk": "ok", "code_len": 2884}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7994, "tout": 13, "ts": 1791536560.7032, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8031, "tout": 1392, "ts": 1791536572.1667762, "chk": null, "code_len": 2884}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791536526.645923}
{"ep": "h2-00949", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0637, "speedup": 1.0637, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 359.2, "calls": 2, "tokens_in": 4933, "tokens_out": 3559, "ctx_chars": 0, "t_end": 1791536706.5258548, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0637, "lo95": 1.0633, "mean": 1.0636, "ev_only": 1.0644, "ref_ms": 8.45933, "cand_ms": 7.95264, "n_trials": 171, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0058, "r_wall_s": 18.3, "compile_s": 1.98, "r_error": "", "src_len": 2373, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 2280, "ts": 1791536366.7476857, "chk": "ok", "code_len": 2373}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3298, "tout": 1279, "ts": 1791536687.929965, "chk": null, "code_len": 2373}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791536347.364156}
{"ep": "h2-00968", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.8736, "speedup": 0.8736, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 304.9, "calls": 3, "tokens_in": 8116, "tokens_out": 2467, "ctx_chars": 0, "t_end": 1791536707.0261188, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8736, "lo95": 0.8725, "mean": 0.8739, "ev_only": 0.8738, "ref_ms": 27.60634, "cand_ms": 31.60188, "n_trials": 40, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.1552, "r_wall_s": 18.07, "compile_s": 2.08, "r_error": "", "src_len": 2073, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1728, "tout": 1246, "ts": 1791536414.4270349, "chk": "ok", "code_len": 2103}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3086, "tout": 192, "ts": 1791536680.034798, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 3302, "tout": 1029, "ts": 1791536688.7002373, "chk": null, "code_len": 2073}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1728, "started": 1791536402.124374}
{"ep": "h2-00962", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.165, "speedup": 1.165, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 328.1, "calls": 3, "tokens_in": 16524, "tokens_out": 12290, "ctx_chars": 0, "t_end": 1791536708.9675758, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.165, "lo95": 1.1642, "mean": 1.1735, "ev_only": 1.1654, "ref_ms": 7.7597, "cand_ms": 6.66066, "n_trials": 147, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 124.86, "compile_s": 12.59, "r_error": "", "src_len": 5981, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 5902, "ts": 1791536430.0795274, "chk": "no_custom_kernel", "code_len": 6052}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5612, "tout": 3459, "ts": 1791536491.449435, "chk": "ok", "code_len": 5981}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8753, "tout": 2929, "ts": 1791536583.4378457, "chk": null, "code_len": 5981}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791536380.893753}
{"ep": "h2-01050", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7707, "speedup": 0.7707, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 54.1, "calls": 3, "tokens_in": 18402, "tokens_out": 1371, "ctx_chars": 8224, "t_end": 1791536716.3279436, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.7707, "lo95": 0.7692, "mean": 0.7707, "ev_only": 0.7705, "ref_ms": 31.43264, "cand_ms": 40.78571, "n_trials": 39, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 13.37, "compile_s": 1.85, "r_error": "", "src_len": 1046, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 768, "ts": 1791536687.8289149, "chk": "ok", "code_len": 1046}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6385, "tout": 100, "ts": 1791536698.0637524, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6509, "tout": 503, "ts": 1791536702.8485274, "chk": null, "code_len": 1046}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791536662.2250912}
{"ep": "h2-00861", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.3213, "speedup": 2.3213, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 646.2, "calls": 2, "tokens_in": 7586, "tokens_out": 11628, "ctx_chars": 0, "t_end": 1791536720.6286602, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 2.3213, "lo95": 2.3207, "mean": 2.3202, "ev_only": 2.3207, "ref_ms": 42.36653, "cand_ms": 18.25099, "n_trials": 42, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9998, "r_wall_s": 312.31, "compile_s": 20.18, "r_error": "", "src_len": 5824, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 8709, "ts": 1791536158.5361068, "chk": "ok", "code_len": 5824}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5550, "tout": 2919, "ts": 1791536407.6400428, "chk": null, "code_len": 5824}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2036, "started": 1791536074.4574783}
{"ep": "h2-01024", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3013, "speedup": 1.3013, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 129.6, "calls": 2, "tokens_in": 14899, "tokens_out": 3818, "ctx_chars": 11065, "t_end": 1791536721.7651637, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3013, "lo95": 1.3, "mean": 1.2995, "ev_only": 1.2918, "ref_ms": 1.80874, "cand_ms": 1.38992, "n_trials": 181, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0639, "r_wall_s": 83.44, "compile_s": 11.16, "r_error": "", "src_len": 2227, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6629, "tout": 2649, "ts": 1791536617.1310072, "chk": "ok", "code_len": 2227}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8270, "tout": 1169, "ts": 1791536638.3012753, "chk": null, "code_len": 2227}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6629, "started": 1791536592.1859336}
{"ep": "h2-01019", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0102, "speedup": 1.0102, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 158.5, "calls": 3, "tokens_in": 24428, "tokens_out": 3817, "ctx_chars": 10167, "t_end": 1791536729.1130602, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0102, "lo95": 1.0085, "mean": 1.0093, "ev_only": 1.009, "ref_ms": 8.4872, "cand_ms": 8.40183, "n_trials": 158, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2624, "r_wall_s": 96.2, "compile_s": 6.56, "r_error": "", "src_len": 1815, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 2903, "ts": 1791536597.7075608, "chk": "ok", "code_len": 1815}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9119, "tout": 62, "ts": 1791536625.1835878, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9205, "tout": 852, "ts": 1791536632.7662702, "chk": null, "code_len": 1815}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791536570.6478229}
{"ep": "h2-00966", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0596, "speedup": 1.0596, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 329.9, "calls": 2, "tokens_in": 14729, "tokens_out": 2427, "ctx_chars": 10527, "t_end": 1791536729.658368, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0596, "lo95": 1.0566, "mean": 1.0583, "ev_only": 1.0579, "ref_ms": 19.78401, "cand_ms": 18.67165, "n_trials": 55, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0103, "r_wall_s": 108.49, "compile_s": 2.93, "r_error": "", "src_len": 2304, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1295, "ts": 1791536411.9905019, "chk": "ok", "code_len": 2423}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8068, "tout": 1132, "ts": 1791536621.1204486, "chk": null, "code_len": 2304}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6661, "started": 1791536399.7498739}
{"ep": "h2-01026", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.857, "speedup": 0.857, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 149.1, "calls": 2, "tokens_in": 14113, "tokens_out": 12657, "ctx_chars": 6513, "t_end": 1791536744.5461733, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.857, "lo95": 0.8568, "mean": 0.8728, "ev_only": 0.8571, "ref_ms": 15.71562, "cand_ms": 18.33764, "n_trials": 74, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.6159, "r_wall_s": 29.32, "compile_s": 1.92, "r_error": "", "src_len": 7595, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 8926, "ts": 1791536673.7025723, "chk": "ok", "code_len": 7648}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9249, "tout": 3731, "ts": 1791536715.0362248, "chk": null, "code_len": 7595}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791536595.3964198}
{"ep": "h2-01013", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.6187, "speedup": 2.6187, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 214.5, "calls": 3, "tokens_in": 16948, "tokens_out": 15387, "ctx_chars": 0, "t_end": 1791536745.3365264, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.6187, "lo95": 2.6036, "mean": 2.5737, "ev_only": 2.6134, "ref_ms": 8.59006, "cand_ms": 3.28027, "n_trials": 190, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9984, "r_wall_s": 28.78, "compile_s": 13.55, "r_error": "", "src_len": 7014, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 7970, "ts": 1791536600.9933589, "chk": "no_custom_kernel", "code_len": 7057}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5739, "tout": 3923, "ts": 1791536677.9358354, "chk": "ok", "code_len": 7014}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9389, "tout": 3494, "ts": 1791536716.524887, "chk": null, "code_len": 7014}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1820, "started": 1791536530.8539963}
{"ep": "h2-01053", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7475, "speedup": 0.7475, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 63.1, "calls": 2, "tokens_in": 14161, "tokens_out": 1798, "ctx_chars": 10850, "t_end": 1791536751.66107, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.7475, "lo95": 0.6976, "mean": 0.673, "ev_only": 0.7237, "ref_ms": 0.1763, "cand_ms": 0.23584, "n_trials": 222, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 35.87, "compile_s": 2.13, "r_error": "", "src_len": 1060, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6624, "tout": 1208, "ts": 1791536700.4622347, "chk": "ok", "code_len": 1060}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7537, "tout": 590, "ts": 1791536715.6456556, "chk": null, "code_len": 1060}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6624, "started": 1791536688.5464709}
{"ep": "h2-00967", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0098, "speedup": 1.0098, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 351.4, "calls": 2, "tokens_in": 16036, "tokens_out": 2176, "ctx_chars": 12117, "t_end": 1791536751.747902, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0098, "lo95": 1.0087, "mean": 1.01, "ev_only": 1.0098, "ref_ms": 42.56242, "cand_ms": 42.14788, "n_trials": 28, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 63.13, "compile_s": 1.93, "r_error": "", "src_len": 2223, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1130, "ts": 1791536410.655452, "chk": "ok", "code_len": 2223}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8639, "tout": 1046, "ts": 1791536687.897267, "chk": null, "code_len": 2223}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7397, "started": 1791536400.3391056}
{"ep": "h2-01023", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.018, "speedup": 1.018, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 163.3, "calls": 3, "tokens_in": 24800, "tokens_out": 2241, "ctx_chars": 12117, "t_end": 1791536754.3963068, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.018, "lo95": 1.0167, "mean": 1.018, "ev_only": 1.0178, "ref_ms": 42.76374, "cand_ms": 42.00603, "n_trials": 31, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 127.67, "compile_s": 3.59, "r_error": "", "src_len": 2244, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1153, "ts": 1791536601.1566525, "chk": "ok", "code_len": 2244}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8662, "tout": 55, "ts": 1791536616.931479, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8741, "tout": 1033, "ts": 1791536625.7092366, "chk": null, "code_len": 2244}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791536591.0547326}
{"ep": "h2-01027", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0033, "speedup": 1.0033, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 159.1, "calls": 3, "tokens_in": 24680, "tokens_out": 2184, "ctx_chars": 12117, "t_end": 1791536755.344268, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0033, "lo95": 0.9997, "mean": 1.0027, "ev_only": 1.0038, "ref_ms": 42.58496, "cand_ms": 42.44674, "n_trials": 29, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 63.54, "compile_s": 2.02, "r_error": "", "src_len": 2387, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1114, "ts": 1791536606.2800522, "chk": "ok", "code_len": 2387}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8623, "tout": 13, "ts": 1791536682.2503772, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8660, "tout": 1057, "ts": 1791536691.1402745, "chk": null, "code_len": 2387}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791536596.254746}
{"ep": "h2-01060", "task": "L3_037", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8042, "speedup": 0.8042, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 53.1, "calls": 3, "tokens_in": 18724, "tokens_out": 1518, "ctx_chars": 8224, "t_end": 1791536759.6751497, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.8042, "lo95": 0.8027, "mean": 0.8123, "ev_only": 0.8042, "ref_ms": 31.98028, "cand_ms": 39.76526, "n_trials": 39, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0003, "r_wall_s": 14.31, "compile_s": 1.96, "r_error": "", "src_len": 1050, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5508, "tout": 945, "ts": 1791536717.2606096, "chk": "ok", "code_len": 1050}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6562, "tout": 68, "ts": 1791536738.4376283, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6654, "tout": 505, "ts": 1791536745.1115687, "chk": null, "code_len": 1050}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5508, "started": 1791536706.5261178}
{"ep": "h2-00999", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.4885, "speedup": 1.4885, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 263.4, "calls": 2, "tokens_in": 15273, "tokens_out": 10162, "ctx_chars": 6513, "t_end": 1791536761.64302, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.4885, "lo95": 1.4849, "mean": 1.473, "ev_only": 1.4868, "ref_ms": 3.32621, "cand_ms": 2.23467, "n_trials": 186, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1454, "r_wall_s": 58.23, "compile_s": 39.52, "r_error": "", "src_len": 8051, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 6324, "ts": 1791536550.0442517, "chk": "ok", "code_len": 8051}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9825, "tout": 3838, "ts": 1791536703.160617, "chk": null, "code_len": 8051}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791536498.285525}
{"ep": "h2-01064", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0583, "speedup": 1.0583, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 43.0, "calls": 3, "tokens_in": 22693, "tokens_out": 2344, "ctx_chars": 10527, "t_end": 1791536763.6591609, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0583, "lo95": 1.0581, "mean": 1.0672, "ev_only": 1.0586, "ref_ms": 19.66455, "cand_ms": 18.58086, "n_trials": 69, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.009, "r_wall_s": 13.36, "compile_s": 1.97, "r_error": "", "src_len": 2209, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1203, "ts": 1791536731.829158, "chk": "ok", "code_len": 2253}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7976, "tout": 56, "ts": 1791536740.7743132, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8056, "tout": 1085, "ts": 1791536750.1810644, "chk": null, "code_len": 2209}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791536720.6291335}
{"ep": "h2-01048", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0433, "speedup": 1.0433, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 118.9, "calls": 2, "tokens_in": 13012, "tokens_out": 2272, "ctx_chars": 9274, "t_end": 1791536764.0350463, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0433, "lo95": 1.043, "mean": 1.05, "ev_only": 1.0439, "ref_ms": 8.28821, "cand_ms": 7.94453, "n_trials": 162, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0057, "r_wall_s": 16.82, "compile_s": 1.84, "r_error": "", "src_len": 1767, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5792, "tout": 1316, "ts": 1791536655.9418454, "chk": "ok", "code_len": 1767}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7220, "tout": 956, "ts": 1791536747.1198711, "chk": null, "code_len": 1767}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5792, "started": 1791536645.146552}
{"ep": "h2-00970", "task": "L3_020", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.3824, "speedup": 1.3824, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 364.9, "calls": 2, "tokens_in": 12201, "tokens_out": 20432, "ctx_chars": 0, "t_end": 1791536771.3070529, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3824, "lo95": 1.3583, "mean": 1.2896, "ev_only": 1.3776, "ref_ms": 7.56372, "cand_ms": 5.47142, "n_trials": 109, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9927, "r_wall_s": 141.07, "compile_s": 4.84, "r_error": "", "src_len": 11499, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2944, "tout": 14643, "ts": 1791536540.3414764, "chk": "ok", "code_len": 11548}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9257, "tout": 5789, "ts": 1791536630.1930747, "chk": null, "code_len": 11499}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2944, "started": 1791536406.4084089}
{"ep": "h2-01058", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.043, "speedup": 1.043, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 78.8, "calls": 2, "tokens_in": 5111, "tokens_out": 3055, "ctx_chars": 0, "t_end": 1791536776.7531028, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.043, "lo95": 1.0428, "mean": 1.0749, "ev_only": 1.0437, "ref_ms": 8.30342, "cand_ms": 7.96071, "n_trials": 167, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0055, "r_wall_s": 28.59, "compile_s": 1.9, "r_error": "", "src_len": 2000, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 1999, "ts": 1791536715.3157685, "chk": "ok", "code_len": 2000}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3345, "tout": 1056, "ts": 1791536748.087783, "chk": null, "code_len": 2000}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791536697.9046817}
{"ep": "h2-01061", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3338, "speedup": 1.3338, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 72.5, "calls": 1, "tokens_in": 5998, "tokens_out": 721, "ctx_chars": 9970, "t_end": 1791536779.5375066, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.3338, "lo95": 1.3325, "mean": 1.3486, "ev_only": 1.3357, "ref_ms": 7.84765, "cand_ms": 5.88384, "n_trials": 172, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0902, "r_wall_s": 64.94, "compile_s": 1.98, "r_error": "", "src_len": 1450, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 721, "ts": 1791536713.9112527, "chk": null, "code_len": 1450}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791536707.0263572}
{"ep": "h2-01054", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0178, "speedup": 1.0178, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 91.3, "calls": 2, "tokens_in": 18019, "tokens_out": 5632, "ctx_chars": 11065, "t_end": 1791536781.7872584, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0178, "lo95": 1.0154, "mean": 1.0582, "ev_only": 1.023, "ref_ms": 1.81036, "cand_ms": 1.77873, "n_trials": 217, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.274, "r_wall_s": 30.53, "compile_s": 1.97, "r_error": "", "src_len": 1971, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 4653, "ts": 1791536733.4785237, "chk": "ok", "code_len": 1971}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 11392, "tout": 979, "ts": 1791536750.9987085, "chk": null, "code_len": 1971}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791536690.4652927}
{"ep": "h2-01029", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 3.696, "speedup": 3.696, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 177.5, "calls": 3, "tokens_in": 15822, "tokens_out": 13711, "ctx_chars": 0, "t_end": 1791536782.3413045, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 3.696, "lo95": 3.6859, "mean": 3.6955, "ev_only": 3.6944, "ref_ms": 59.62012, "cand_ms": 16.13078, "n_trials": 23, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8447, "r_wall_s": 21.18, "compile_s": 11.12, "r_error": "", "src_len": 5883, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 7888, "ts": 1791536671.7746675, "chk": "no_custom_kernel", "code_len": 5905}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5437, "tout": 2968, "ts": 1791536707.473281, "chk": "ok", "code_len": 5883}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8399, "tout": 2855, "ts": 1791536761.1028316, "chk": null, "code_len": 5883}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791536604.851719}
{"ep": "h2-01018", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0105, "speedup": 1.0105, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 225.2, "calls": 2, "tokens_in": 6377, "tokens_out": 7605, "ctx_chars": 0, "t_end": 1791536783.936564, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0105, "lo95": 1.0102, "mean": 1.0106, "ev_only": 1.0106, "ref_ms": 30.0407, "cand_ms": 29.72822, "n_trials": 41, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0171, "r_wall_s": 141.96, "compile_s": 11.1, "r_error": "", "src_len": 5061, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1598, "tout": 4969, "ts": 1791536601.6881905, "chk": "ok", "code_len": 5061}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4779, "tout": 2636, "ts": 1791536641.5069954, "chk": null, "code_len": 5061}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1598, "started": 1791536558.693979}
{"ep": "h2-01059", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0601, "speedup": 1.0601, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.8, "calls": 3, "tokens_in": 25683, "tokens_out": 4551, "ctx_chars": 10167, "t_end": 1791536787.3435004, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0601, "lo95": 1.0595, "mean": 1.0614, "ev_only": 1.0612, "ref_ms": 8.44107, "cand_ms": 7.96228, "n_trials": 149, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0037, "r_wall_s": 25.07, "compile_s": 2.09, "r_error": "", "src_len": 1913, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3530, "ts": 1791536743.4461422, "chk": "ok", "code_len": 1913}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9746, "tout": 63, "ts": 1791536753.7955813, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9833, "tout": 958, "ts": 1791536762.1827111, "chk": null, "code_len": 1913}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6104, "started": 1791536705.57457}
{"ep": "h2-01071", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.847, "speedup": 0.847, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 36.8, "calls": 1, "tokens_in": 7030, "tokens_out": 1232, "ctx_chars": 11175, "t_end": 1791536788.4986482, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.847, "lo95": 0.8464, "mean": 0.8479, "ev_only": 0.8465, "ref_ms": 14.31711, "cand_ms": 16.90248, "n_trials": 99, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 25.54, "compile_s": 2.04, "r_error": "", "src_len": 2621, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 7030, "tout": 1232, "ts": 1791536762.6328144, "chk": null, "code_len": 2621}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 7030, "started": 1791536751.748127}
{"ep": "h2-01056", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0188, "speedup": 1.0188, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.6, "calls": 2, "tokens_in": 5108, "tokens_out": 4344, "ctx_chars": 0, "t_end": 1791536788.8571823, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0188, "lo95": 1.0174, "mean": 1.019, "ev_only": 1.0186, "ref_ms": 27.64872, "cand_ms": 27.13829, "n_trials": 51, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0136, "r_wall_s": 36.33, "compile_s": 2.35, "r_error": "", "src_len": 2169, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 3156, "ts": 1791536722.1466649, "chk": "ok", "code_len": 2169}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3378, "tout": 1188, "ts": 1791536752.2805836, "chk": null, "code_len": 2169}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791536691.3071191}
{"ep": "h2-01028", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0282, "speedup": 1.0282, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 191.5, "calls": 2, "tokens_in": 9175, "tokens_out": 2417, "ctx_chars": 5026, "t_end": 1791536789.7546446, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0282, "lo95": 1.025, "mean": 1.0366, "ev_only": 1.026, "ref_ms": 8.22558, "cand_ms": 8.00009, "n_trials": 151, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0066, "r_wall_s": 159.14, "compile_s": 8.37, "r_error": "", "src_len": 1289, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1678, "ts": 1791536613.0300732, "chk": "ok", "code_len": 1289}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5180, "tout": 739, "ts": 1791536630.5377212, "chk": null, "code_len": 1289}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791536598.2062018}
{"ep": "h2-01055", "task": "L3_009", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.3239, "speedup": 0.3239, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 121.1, "calls": 4, "tokens_in": 37495, "tokens_out": 9033, "ctx_chars": 6513, "t_end": 1791536812.1336586, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.3239, "lo95": 0.3224, "mean": 0.3137, "ev_only": 0.3227, "ref_ms": 3.25214, "cand_ms": 10.04098, "n_trials": 162, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0249, "r_wall_s": 37.81, "compile_s": 1.98, "r_error": "", "src_len": 6123, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5446, "tout": 3055, "ts": 1791536716.398218, "chk": "forbidden", "code_len": 6158}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8603, "tout": 2964, "ts": 1791536739.984437, "chk": "ok", "code_len": 6123}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11679, "tout": 64, "ts": 1791536750.7280233, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11767, "tout": 2950, "ts": 1791536774.2105012, "chk": null, "code_len": 6123}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "ok", "no_code", null], "first_tin": 5446, "started": 1791536691.0450819}
{"ep": "h2-01022", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.5893, "speedup": 4.5893, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 226.1, "calls": 3, "tokens_in": 16578, "tokens_out": 16527, "ctx_chars": 0, "t_end": 1791536813.5169983, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 4.5893, "lo95": 4.5786, "mean": 4.6268, "ev_only": 4.6658, "ref_ms": 8.53028, "cand_ms": 1.85874, "n_trials": 176, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9988, "r_wall_s": 54.09, "compile_s": 11.84, "r_error": "", "src_len": 6717, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 9434, "ts": 1791536674.2448661, "chk": "no_custom_kernel", "code_len": 6741}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5620, "tout": 3817, "ts": 1791536716.6672168, "chk": "ok", "code_len": 6717}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9138, "tout": 3276, "ts": 1791536759.3842723, "chk": null, "code_len": 6717}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1820, "started": 1791536587.3876143}
{"ep": "h2-01052", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0166, "speedup": 1.0166, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 145.8, "calls": 5, "tokens_in": 50062, "tokens_out": 5716, "ctx_chars": 11823, "t_end": 1791536814.8650744, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0166, "lo95": 1.015, "mean": 1.0167, "ev_only": 1.0169, "ref_ms": 27.54855, "cand_ms": 27.09942, "n_trials": 49, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0151, "r_wall_s": 44.12, "compile_s": 1.85, "r_error": "", "src_len": 2005, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1481, "ts": 1791536699.6734922, "chk": "incorrect", "code_len": 2663}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8534, "tout": 1993, "ts": 1791536727.8630693, "chk": "ok", "code_len": 2482}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 10639, "tout": 1195, "ts": 1791536748.202359, "chk": "ok", "code_len": 2005}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11946, "tout": 29, "ts": 1791536758.0550578, "chk": "no_code", "code_len": 0}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 12022, "tout": 1018, "ts": 1791536770.6396813, "chk": null, "code_len": 2005}], "n_turns": 5, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "ok", "no_code", null], "first_tin": 6921, "started": 1791536669.7309263}
{"ep": "h2-01049", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.3361, "speedup": 1.3361, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 171.6, "calls": 1, "tokens_in": 5998, "tokens_out": 718, "ctx_chars": 9970, "t_end": 1791536816.9429781, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3361, "lo95": 1.3354, "mean": 1.339, "ev_only": 1.3337, "ref_ms": 7.81618, "cand_ms": 5.85001, "n_trials": 187, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0922, "r_wall_s": 164.06, "compile_s": 23.1, "r_error": "", "src_len": 1374, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 718, "ts": 1791536652.1850877, "chk": null, "code_len": 1374}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791536645.345238}
{"ep": "h2-01075", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.5537, "speedup": 0.5537, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 56.7, "calls": 1, "tokens_in": 1565, "tokens_out": 1779, "ctx_chars": 0, "t_end": 1791536818.3511734, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.5537, "lo95": 0.5536, "mean": 0.5537, "ev_only": 0.5537, "ref_ms": 15.46973, "cand_ms": 27.9368, "n_trials": 52, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 41.05, "compile_s": 2.05, "r_error": "", "src_len": 3055, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1565, "tout": 1779, "ts": 1791536776.8929355, "chk": null, "code_len": 3055}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1565, "started": 1791536761.6432612}
{"ep": "h2-01079", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0234, "speedup": 1.0234, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 48.8, "calls": 2, "tokens_in": 9161, "tokens_out": 2271, "ctx_chars": 5026, "t_end": 1791536825.6002285, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0234, "lo95": 1.0232, "mean": 1.0278, "ev_only": 1.0227, "ref_ms": 8.13673, "cand_ms": 7.95077, "n_trials": 142, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0067, "r_wall_s": 18.76, "compile_s": 1.96, "r_error": "", "src_len": 1304, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 3995, "tout": 1568, "ts": 1791536793.0950842, "chk": "ok", "code_len": 1304}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5166, "tout": 703, "ts": 1791536806.7446547, "chk": null, "code_len": 1304}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 3995, "started": 1791536776.7533658}
{"ep": "h2-00955", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0624, "speedup": 1.0624, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 471.3, "calls": 3, "tokens_in": 8409, "tokens_out": 2176, "ctx_chars": 0, "t_end": 1791536830.8797078, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0624, "lo95": 1.0602, "mean": 1.0621, "ev_only": 1.0624, "ref_ms": 19.75598, "cand_ms": 18.5951, "n_trials": 58, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0084, "r_wall_s": 180.13, "compile_s": 21.62, "r_error": "", "src_len": 2125, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1956, "tout": 1140, "ts": 1791536370.7374458, "chk": "ok", "code_len": 2241}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3208, "tout": 13, "ts": 1791536642.4835713, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 3245, "tout": 1023, "ts": 1791536650.700628, "chk": null, "code_len": 2125}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1956, "started": 1791536359.6021137}
{"ep": "h2-00989", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 8.8909, "speedup": 8.8909, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 369.8, "calls": 4, "tokens_in": 25683, "tokens_out": 19751, "ctx_chars": 0, "t_end": 1791536833.8623111, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 8.8909, "lo95": 8.8867, "mean": 8.8826, "ev_only": 8.9169, "ref_ms": 64.37019, "cand_ms": 7.24001, "n_trials": 23, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.2287, "r_wall_s": 20.51, "compile_s": 11.58, "r_error": "", "src_len": 6267, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 10143, "ts": 1791536557.2513115, "chk": "runtime_error", "code_len": 4694}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5025, "tout": 2840, "ts": 1791536591.8497097, "chk": "no_custom_kernel", "code_len": 4675}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 7772, "tout": 3654, "ts": 1791536634.4365447, "chk": "ok", "code_len": 6267}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11140, "tout": 3114, "ts": 1791536813.3041177, "chk": null, "code_len": 6267}], "n_turns": 4, "n_check": 3, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["runtime_error", "no_custom_kernel", "ok", null], "first_tin": 1746, "started": 1791536464.0915504}
{"ep": "h2-01066", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3217, "speedup": 1.3217, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 107.4, "calls": 2, "tokens_in": 13178, "tokens_out": 2601, "ctx_chars": 9970, "t_end": 1791536836.5625432, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.3217, "lo95": 1.3213, "mean": 1.3214, "ev_only": 1.3203, "ref_ms": 7.80718, "cand_ms": 5.90681, "n_trials": 217, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0955, "r_wall_s": 59.23, "compile_s": 1.93, "r_error": "", "src_len": 1378, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 1861, "ts": 1791536752.421169, "chk": "ok", "code_len": 1378}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7178, "tout": 740, "ts": 1791536776.6521025, "chk": null, "code_len": 1378}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791536729.644439}
{"ep": "h2-01062", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.1269, "speedup": 2.1269, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 131.7, "calls": 2, "tokens_in": 7600, "tokens_out": 8579, "ctx_chars": 0, "t_end": 1791536840.67944, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.1269, "lo95": 2.1256, "mean": 2.1267, "ev_only": 2.1302, "ref_ms": 14.28824, "cand_ms": 6.71784, "n_trials": 102, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 29.49, "compile_s": 11.41, "r_error": "", "src_len": 5473, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 5998, "ts": 1791536762.265734, "chk": "ok", "code_len": 5473}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5417, "tout": 2581, "ts": 1791536811.058948, "chk": null, "code_len": 5473}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2183, "started": 1791536708.967857}
{"ep": "h2-01035", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.7163, "speedup": 0.7163, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 219.2, "calls": 2, "tokens_in": 16477, "tokens_out": 4150, "ctx_chars": 12352, "t_end": 1791536845.18182, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.7163, "lo95": 0.7147, "mean": 0.7165, "ev_only": 0.7159, "ref_ms": 64.39882, "cand_ms": 89.90026, "n_trials": 16, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0019, "r_wall_s": 36.32, "compile_s": 2.27, "r_error": "", "src_len": 2498, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7330, "tout": 2841, "ts": 1791536650.736492, "chk": "ok", "code_len": 2498}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9147, "tout": 1309, "ts": 1791536808.815822, "chk": null, "code_len": 2498}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7330, "started": 1791536625.9835422}
{"ep": "h2-01083", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "incorrect", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 63.1, "calls": 1, "tokens_in": 2942, "tokens_out": 4654, "ctx_chars": 0, "t_end": 1791536847.0748346, "error": "output mismatch beyond atol=rtol=0.0001", "f_status": "incorrect", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "triton", "own_frac": 0.1131, "r_wall_s": 9.13, "compile_s": 2.73, "r_error": "output mismatch beyond atol=rtol=0.0001", "src_len": 5285, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 2942, "tout": 4654, "ts": 1791536837.9021494, "chk": null, "code_len": 5285}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 2942, "started": 1791536783.9372144}
{"ep": "h2-01030", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4837, "speedup": 1.4837, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 246.4, "calls": 2, "tokens_in": 10544, "tokens_out": 20380, "ctx_chars": 0, "t_end": 1791536853.804217, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.4837, "lo95": 1.4816, "mean": 1.492, "ev_only": 1.4942, "ref_ms": 15.80105, "cand_ms": 10.64982, "n_trials": 86, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1592, "r_wall_s": 62.48, "compile_s": 11.79, "r_error": "", "src_len": 11446, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2161, "tout": 14675, "ts": 1791536735.9639435, "chk": "ok", "code_len": 11559}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8383, "tout": 5705, "ts": 1791536790.989226, "chk": null, "code_len": 11446}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2161, "started": 1791536607.9988308}
{"ep": "h2-01038", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.3345, "speedup": 1.3345, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 235.5, "calls": 1, "tokens_in": 5998, "tokens_out": 745, "ctx_chars": 9970, "t_end": 1791536864.8626719, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.3345, "lo95": 1.334, "mean": 1.334, "ev_only": 1.3323, "ref_ms": 7.79948, "cand_ms": 5.84461, "n_trials": 175, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1109, "r_wall_s": 227.82, "compile_s": 9.7, "r_error": "", "src_len": 1433, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 745, "ts": 1791536636.2827256, "chk": null, "code_len": 1433}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791536629.4104238}
{"ep": "h2-01084", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0277, "speedup": 1.0277, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 88.1, "calls": 1, "tokens_in": 5832, "tokens_out": 524, "ctx_chars": 9656, "t_end": 1791536875.4621215, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.0277, "lo95": 1.0267, "mean": 1.0277, "ev_only": 1.028, "ref_ms": 24.92156, "cand_ms": 24.25033, "n_trials": 44, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0224, "r_wall_s": 73.51, "compile_s": 1.98, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 524, "ts": 1791536801.1820757, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791536787.3437357}
{"ep": "h2-01051", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8062, "speedup": 0.8062, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 208.5, "calls": 5, "tokens_in": 58243, "tokens_out": 13129, "ctx_chars": 6513, "t_end": 1791536875.652286, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8062, "lo95": 0.7937, "mean": 0.7942, "ev_only": 0.8057, "ref_ms": 7.54329, "cand_ms": 9.357, "n_trials": 170, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0007, "r_wall_s": 27.81, "compile_s": 2.12, "r_error": "", "src_len": 3775, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2699, "ts": 1791536699.065019, "chk": "incorrect", "code_len": 6149}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8476, "tout": 3077, "ts": 1791536735.9942899, "chk": "incorrect", "code_len": 6475}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 11685, "tout": 3270, "ts": 1791536773.6174664, "chk": "incorrect", "code_len": 6421}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 15087, "tout": 2131, "ts": 1791536818.8623457, "chk": "ok", "code_len": 3775}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 17350, "tout": 1952, "ts": 1791536847.7995384, "chk": null, "code_len": 3775}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "incorrect", "ok", null], "first_tin": 5645, "started": 1791536667.1118596}
{"ep": "h2-01067", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0257, "speedup": 1.0257, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 151.3, "calls": 2, "tokens_in": 4654, "tokens_out": 2508, "ctx_chars": 0, "t_end": 1791536880.9763033, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0257, "lo95": 1.0246, "mean": 1.0258, "ev_only": 1.0256, "ref_ms": 8.23147, "cand_ms": 8.02515, "n_trials": 153, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0087, "r_wall_s": 59.84, "compile_s": 2.63, "r_error": "", "src_len": 1336, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 1715, "ts": 1791536763.149509, "chk": "ok", "code_len": 1336}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2902, "tout": 793, "ts": 1791536821.0613039, "chk": null, "code_len": 1336}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1752, "started": 1791536729.6585882}
{"ep": "h2-01039", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 2.798, "speedup": 2.798, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.0, "calls": 2, "tokens_in": 13979, "tokens_out": 4238, "ctx_chars": 9970, "t_end": 1791536882.5321996, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.798, "lo95": 2.7967, "mean": 2.8498, "ev_only": 2.8086, "ref_ms": 7.80834, "cand_ms": 2.79065, "n_trials": 157, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 57.87, "compile_s": 2.14, "r_error": "", "src_len": 2620, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6000, "tout": 2862, "ts": 1791536655.2511778, "chk": "ok", "code_len": 2620}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7979, "tout": 1376, "ts": 1791536823.9867382, "chk": null, "code_len": 2620}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6000, "started": 1791536630.49082}
{"ep": "h2-01087", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.057, "speedup": 1.057, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 95.2, "calls": 3, "tokens_in": 22730, "tokens_out": 2371, "ctx_chars": 10527, "t_end": 1791536885.0015204, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.057, "lo95": 1.0564, "mean": 1.0569, "ev_only": 1.0563, "ref_ms": 19.5839, "cand_ms": 18.52788, "n_trials": 61, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.009, "r_wall_s": 14.36, "compile_s": 1.92, "r_error": "", "src_len": 2308, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1243, "ts": 1791536835.7295887, "chk": "ok", "code_len": 2308}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8016, "tout": 13, "ts": 1791536861.1806393, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8053, "tout": 1115, "ts": 1791536870.5258498, "chk": null, "code_len": 2308}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6661, "started": 1791536790.2827137}
{"ep": "h2-00896", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.7488, "speedup": 0.7488, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 729.1, "calls": 5, "tokens_in": 53635, "tokens_out": 15514, "ctx_chars": 0, "t_end": 1791536888.6180358, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7488, "lo95": 0.6332, "mean": 0.7862, "ev_only": 0.7485, "ref_ms": 9.87003, "cand_ms": 13.18184, "n_trials": 121, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.1088, "r_wall_s": 261.3, "compile_s": 21.03, "r_error": "", "src_len": 4186, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2942, "tout": 3654, "ts": 1791536192.0518436, "chk": "forbidden", "code_len": 7831}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6736, "tout": 3266, "ts": 1791536227.4902756, "chk": "compile_error", "code_len": 7173}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 11217, "tout": 3253, "ts": 1791536304.7244604, "chk": "incorrect", "code_len": 7207}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 14601, "tout": 3406, "ts": 1791536367.0503144, "chk": "ok", "code_len": 4186}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 18139, "tout": 1935, "ts": 1791536627.2871866, "chk": null, "code_len": 4186}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["forbidden", "compile_error", "incorrect", "ok", null], "first_tin": 2942, "started": 1791536159.50978}
{"ep": "h2-01091", "task": "L2_086", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0089, "speedup": 1.0089, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 73.4, "calls": 2, "tokens_in": 4283, "tokens_out": 2699, "ctx_chars": 0, "t_end": 1791536890.365608, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0089, "lo95": 1.0084, "mean": 1.0119, "ev_only": 1.0113, "ref_ms": 7.97366, "cand_ms": 7.9036, "n_trials": 146, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0045, "r_wall_s": 25.45, "compile_s": 1.91, "r_error": "", "src_len": 1196, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1601, "tout": 2032, "ts": 1791536836.7032466, "chk": "ok", "code_len": 1196}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2682, "tout": 667, "ts": 1791536864.7570386, "chk": null, "code_len": 1196}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1601, "started": 1791536816.9432201}
{"ep": "h2-01089", "task": "L2_077", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0176, "speedup": 1.0176, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 77.4, "calls": 2, "tokens_in": 16305, "tokens_out": 2551, "ctx_chars": 12352, "t_end": 1791536890.885682, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.0176, "lo95": 1.0172, "mean": 1.0377, "ev_only": 1.0179, "ref_ms": 64.42895, "cand_ms": 63.31618, "n_trials": 20, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0031, "r_wall_s": 21.24, "compile_s": 1.87, "r_error": "", "src_len": 1982, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7328, "tout": 1537, "ts": 1791536828.9424253, "chk": "ok", "code_len": 1982}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8977, "tout": 1014, "ts": 1791536869.625771, "chk": null, "code_len": 1982}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7328, "started": 1791536813.51724}
{"ep": "h2-01069", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.7951, "speedup": 0.7951, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 146.0, "calls": 3, "tokens_in": 18888, "tokens_out": 1691, "ctx_chars": 8224, "t_end": 1791536891.3704097, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 0.7951, "lo95": 0.7913, "mean": 0.817, "ev_only": 0.7951, "ref_ms": 60.72975, "cand_ms": 76.38314, "n_trials": 14, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 74.77, "compile_s": 2.27, "r_error": "", "src_len": 1106, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 902, "ts": 1791536759.2320786, "chk": "ok", "code_len": 1106}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6539, "tout": 258, "ts": 1791536778.415781, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6821, "tout": 531, "ts": 1791536816.5398452, "chk": null, "code_len": 1106}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 5528, "started": 1791536745.3368351}
{"ep": "h2-01044", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 6.2663, "speedup": 6.2663, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 254.4, "calls": 3, "tokens_in": 17138, "tokens_out": 18207, "ctx_chars": 0, "t_end": 1791536892.2746387, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 6.2663, "lo95": 6.2121, "mean": 6.8434, "ev_only": 6.2676, "ref_ms": 32.73795, "cand_ms": 5.22448, "n_trials": 23, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9857, "r_wall_s": 71.65, "compile_s": 12.1, "r_error": "", "src_len": 6629, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 11380, "ts": 1791536738.736103, "chk": "no_custom_kernel", "code_len": 6628}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5882, "tout": 3587, "ts": 1791536779.7118587, "chk": "ok", "code_len": 6629}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9290, "tout": 3240, "ts": 1791536820.581881, "chk": null, "code_len": 6629}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791536637.8882995}
{"ep": "h2-01085", "task": "L1_061", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0087, "speedup": 1.0087, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 109.4, "calls": 3, "tokens_in": 24630, "tokens_out": 2126, "ctx_chars": 12117, "t_end": 1791536897.8880017, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0087, "lo95": 1.0072, "mean": 1.0083, "ev_only": 1.0087, "ref_ms": 42.49105, "cand_ms": 42.12382, "n_trials": 37, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 65.86, "compile_s": 1.9, "r_error": "", "src_len": 2269, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7397, "tout": 1089, "ts": 1791536806.0939307, "chk": "ok", "code_len": 2269}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8598, "tout": 13, "ts": 1791536822.4238932, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8635, "tout": 1024, "ts": 1791536831.3496995, "chk": null, "code_len": 2269}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7397, "started": 1791536788.4989257}
{"ep": "h2-01072", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.1435, "speedup": 1.1435, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 146.1, "calls": 2, "tokens_in": 13786, "tokens_out": 11881, "ctx_chars": 6513, "t_end": 1791536900.5324986, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.1435, "lo95": 1.1409, "mean": 1.1439, "ev_only": 1.1419, "ref_ms": 15.7807, "cand_ms": 13.80016, "n_trials": 91, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.4927, "r_wall_s": 36.85, "compile_s": 1.82, "r_error": "", "src_len": 7294, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 8336, "ts": 1791536829.2235394, "chk": "ok", "code_len": 7294}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8922, "tout": 3545, "ts": 1791536863.3327947, "chk": null, "code_len": 7294}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791536754.3965712}
{"ep": "h2-01047", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.1065, "speedup": 1.1065, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 258.6, "calls": 2, "tokens_in": 15481, "tokens_out": 3409, "ctx_chars": 11065, "t_end": 1791536902.4389129, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.1065, "lo95": 1.103, "mean": 1.0989, "ev_only": 1.1099, "ref_ms": 1.81518, "cand_ms": 1.64054, "n_trials": 218, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1809, "r_wall_s": 32.24, "compile_s": 1.91, "r_error": "", "src_len": 2507, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 2115, "ts": 1791536666.8432338, "chk": "ok", "code_len": 2507}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8854, "tout": 1294, "ts": 1791536869.9368246, "chk": null, "code_len": 2507}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791536643.8145401}
{"ep": "h2-01093", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0029, "speedup": 1.0029, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 79.0, "calls": 3, "tokens_in": 24039, "tokens_out": 2796, "ctx_chars": 11823, "t_end": 1791536904.6129546, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0029, "lo95": 1.0017, "mean": 1.0035, "ev_only": 1.0039, "ref_ms": 27.52723, "cand_ms": 27.44673, "n_trials": 41, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0226, "r_wall_s": 32.62, "compile_s": 2.01, "r_error": "", "src_len": 2517, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1463, "ts": 1791536839.4475865, "chk": "ok", "code_len": 2639}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8496, "tout": 102, "ts": 1791536861.2016988, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8622, "tout": 1231, "ts": 1791536871.7397182, "chk": null, "code_len": 2517}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6921, "started": 1791536825.600479}
{"ep": "h2-01041", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.1115, "speedup": 2.1115, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 279.0, "calls": 3, "tokens_in": 14031, "tokens_out": 10436, "ctx_chars": 0, "t_end": 1791536910.7054443, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 2.1115, "lo95": 2.1093, "mean": 2.1083, "ev_only": 2.1082, "ref_ms": 7.83992, "cand_ms": 3.7129, "n_trials": 208, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9977, "r_wall_s": 97.49, "compile_s": 14.49, "r_error": "", "src_len": 4880, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 5392, "ts": 1791536680.051445, "chk": "no_custom_kernel", "code_len": 4902}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4850, "tout": 2612, "ts": 1791536754.5029674, "chk": "ok", "code_len": 4880}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7456, "tout": 2432, "ts": 1791536812.0491793, "chk": null, "code_len": 4880}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1725, "started": 1791536631.6792247}
{"ep": "h2-00996", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0014, "speedup": 1.0014, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 430.6, "calls": 3, "tokens_in": 9177, "tokens_out": 3992, "ctx_chars": 0, "t_end": 1791536911.8038933, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0014, "lo95": 1.0006, "mean": 1.0009, "ev_only": 1.0027, "ref_ms": 30.06041, "cand_ms": 30.01893, "n_trials": 52, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0287, "r_wall_s": 281.77, "compile_s": 66.2, "r_error": "", "src_len": 4428, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1596, "tout": 2064, "ts": 1791536544.1887379, "chk": "ok", "code_len": 4481}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3772, "tout": 13, "ts": 1791536610.342489, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 3809, "tout": 1915, "ts": 1791536629.4303358, "chk": null, "code_len": 4428}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 1596, "started": 1791536481.7246914}
{"ep": "h2-01074", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.6924, "speedup": 2.6924, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 154.2, "calls": 2, "tokens_in": 5620, "tokens_out": 5236, "ctx_chars": 0, "t_end": 1791536913.8763049, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 2.6924, "lo95": 2.6903, "mean": 2.6915, "ev_only": 2.6872, "ref_ms": 8.57737, "cand_ms": 3.18573, "n_trials": 177, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.3034, "r_wall_s": 101.35, "compile_s": 2.65, "r_error": "", "src_len": 2968, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 3580, "ts": 1791536791.91083, "chk": "ok", "code_len": 2987}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3800, "tout": 1656, "ts": 1791536812.4709854, "chk": null, "code_len": 2968}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791536759.6753898}
{"ep": "h2-01034", "task": "L2_039", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9958, "speedup": 0.9958, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 297.4, "calls": 3, "tokens_in": 22368, "tokens_out": 2929, "ctx_chars": 10823, "t_end": 1791536914.1196458, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9958, "lo95": 0.9951, "mean": 0.9954, "ev_only": 0.9955, "ref_ms": 30.09023, "cand_ms": 30.21686, "n_trials": 48, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0371, "r_wall_s": 37.78, "compile_s": 1.98, "r_error": "", "src_len": 2875, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6353, "tout": 1524, "ts": 1791536630.086081, "chk": "ok", "code_len": 2875}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7989, "tout": 13, "ts": 1791536864.1750941, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8026, "tout": 1392, "ts": 1791536875.714515, "chk": null, "code_len": 2875}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6353, "started": 1791536616.676648}
{"ep": "h2-01078", "task": "L2_089", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 10.0, "speedup": 34.4743, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 144.6, "calls": 2, "tokens_in": 6923, "tokens_out": 6792, "ctx_chars": 0, "t_end": 1791536915.8782003, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 34.4743, "lo95": 34.3696, "mean": 34.0994, "ev_only": 35.4637, "ref_ms": 19.73253, "cand_ms": 0.57238, "n_trials": 64, "fast1": true, "flags": ["excessive_speedup>10.0x: audit"], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 72.1, "compile_s": 18.83, "r_error": "", "src_len": 5100, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1958, "tout": 4388, "ts": 1791536808.2936604, "chk": "ok", "code_len": 5100}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4965, "tout": 2404, "ts": 1791536843.7178066, "chk": null, "code_len": 5100}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1958, "started": 1791536771.3074574}
{"ep": "h2-01046", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.0277, "speedup": 1.0277, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 275.6, "calls": 1, "tokens_in": 5832, "tokens_out": 524, "ctx_chars": 9656, "t_end": 1791536915.878872, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0277, "lo95": 1.0265, "mean": 1.0272, "ev_only": 1.0272, "ref_ms": 24.82365, "cand_ms": 24.15448, "n_trials": 69, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0288, "r_wall_s": 269.56, "compile_s": 21.76, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 524, "ts": 1791536645.2574437, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791536640.2453218}
{"ep": "h2-01065", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.4458, "speedup": 1.4458, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 194.5, "calls": 2, "tokens_in": 11931, "tokens_out": 7040, "ctx_chars": 6513, "t_end": 1791536916.256529, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 1.4458, "lo95": 1.4441, "mean": 1.4406, "ev_only": 1.4458, "ref_ms": 15.56117, "cand_ms": 10.76282, "n_trials": 62, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1859, "r_wall_s": 102.8, "compile_s": 2.63, "r_error": "", "src_len": 4960, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 4736, "ts": 1791536764.2208257, "chk": "ok", "code_len": 4960}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7324, "tout": 2304, "ts": 1791536812.636297, "chk": null, "code_len": 4960}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791536721.7654133}
{"ep": "h2-01002", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 9.2652, "speedup": 9.2652, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 413.2, "calls": 2, "tokens_in": 6794, "tokens_out": 11381, "ctx_chars": 0, "t_end": 1791536917.2384512, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-02", "gpu": 0, "r_speedup": 9.2652, "lo95": 9.2093, "mean": 9.2379, "ev_only": 9.2701, "ref_ms": 64.25582, "cand_ms": 6.93515, "n_trials": 25, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.219, "r_wall_s": 63.75, "compile_s": 17.4, "r_error": "", "src_len": 5160, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 8775, "ts": 1791536641.838426, "chk": "ok", "code_len": 5283}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5048, "tout": 2606, "ts": 1791536853.4033725, "chk": null, "code_len": 5160}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791536504.0608594}
{"ep": "h2-01068", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9247, "speedup": 0.9247, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 188.6, "calls": 3, "tokens_in": 23906, "tokens_out": 1991, "ctx_chars": 11539, "t_end": 1791536933.11607, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 0.9247, "lo95": 0.9235, "mean": 0.9263, "ev_only": 0.924, "ref_ms": 7.87602, "cand_ms": 8.51704, "n_trials": 130, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0602, "r_wall_s": 113.31, "compile_s": 2.68, "r_error": "", "src_len": 1767, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7113, "tout": 1156, "ts": 1791536763.1816924, "chk": "ok", "code_len": 1792}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8378, "tout": 13, "ts": 1791536778.9167776, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8415, "tout": 822, "ts": 1791536818.7105107, "chk": null, "code_len": 1767}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7113, "started": 1791536744.5464265}
{"ep": "h2-00956", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9918, "speedup": 0.9918, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 569.8, "calls": 2, "tokens_in": 16355, "tokens_out": 3969, "ctx_chars": 11065, "t_end": 1791536934.5525415, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.9918, "lo95": 0.9713, "mean": 0.9922, "ev_only": 0.997, "ref_ms": 1.79664, "cand_ms": 1.81141, "n_trials": 209, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.2903, "r_wall_s": 272.14, "compile_s": 16.68, "r_error": "", "src_len": 1976, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6627, "tout": 2989, "ts": 1791536396.2358606, "chk": "ok", "code_len": 1976}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9728, "tout": 980, "ts": 1791536662.383138, "chk": null, "code_len": 1976}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6627, "started": 1791536364.7968135}
{"ep": "h2-01105", "task": "L2_048", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.3305, "speedup": 1.3305, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 59.9, "calls": 1, "tokens_in": 5998, "tokens_out": 764, "ctx_chars": 9970, "t_end": 1791536942.4249368, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-22", "gpu": 0, "r_speedup": 1.3305, "lo95": 1.33, "mean": 1.3367, "ev_only": 1.3321, "ref_ms": 7.81137, "cand_ms": 5.87085, "n_trials": 205, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0903, "r_wall_s": 51.75, "compile_s": 1.89, "r_error": "", "src_len": 1433, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5998, "tout": 764, "ts": 1791536890.0511355, "chk": null, "code_len": 1433}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5998, "started": 1791536882.532449}
{"ep": "h2-01099", "task": "L2_086", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0039, "speedup": 1.0039, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 97.3, "calls": 3, "tokens_in": 20676, "tokens_out": 976, "ctx_chars": 10909, "t_end": 1791536944.3435464, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0039, "lo95": 1.003, "mean": 1.0035, "ev_only": 1.0044, "ref_ms": 7.97883, "cand_ms": 7.94813, "n_trials": 164, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.004, "r_wall_s": 69.07, "compile_s": 2.41, "r_error": "", "src_len": 935, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6477, "tout": 495, "ts": 1791536861.363961, "chk": "ok", "code_len": 962}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7081, "tout": 13, "ts": 1791536870.3460858, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 7118, "tout": 468, "ts": 1791536875.1160235, "chk": null, "code_len": 935}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6477, "started": 1791536847.0750816}
{"ep": "h2-01080", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0233, "speedup": 1.0233, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 165.2, "calls": 1, "tokens_in": 5832, "tokens_out": 524, "ctx_chars": 9656, "t_end": 1791536944.7475696, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0233, "lo95": 1.0221, "mean": 1.0239, "ev_only": 1.0253, "ref_ms": 25.07293, "cand_ms": 24.50217, "n_trials": 53, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0224, "r_wall_s": 126.44, "compile_s": 2.92, "r_error": "", "src_len": 996, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 5832, "tout": 524, "ts": 1791536817.5723553, "chk": null, "code_len": 996}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 5832, "started": 1791536779.5377405}
{"ep": "h2-01042", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0238, "speedup": 1.0238, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 322.4, "calls": 2, "tokens_in": 4750, "tokens_out": 2526, "ctx_chars": 0, "t_end": 1791536955.647258, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0238, "lo95": 1.0229, "mean": 1.0229, "ev_only": 1.0257, "ref_ms": 8.22322, "cand_ms": 8.03237, "n_trials": 161, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.007, "r_wall_s": 109.96, "compile_s": 3.74, "r_error": "", "src_len": 1348, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 1736, "ts": 1791536649.4230723, "chk": "ok", "code_len": 1348}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2998, "tout": 790, "ts": 1791536845.6121519, "chk": null, "code_len": 1348}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1752, "started": 1791536633.2556226}
{"ep": "h2-01021", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0183, "speedup": 1.0183, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 371.6, "calls": 2, "tokens_in": 5664, "tokens_out": 4616, "ctx_chars": 0, "t_end": 1791536956.017959, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0183, "lo95": 1.0157, "mean": 1.0674, "ev_only": 1.0195, "ref_ms": 27.56529, "cand_ms": 27.06934, "n_trials": 28, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0146, "r_wall_s": 312.22, "compile_s": 32.03, "r_error": "", "src_len": 3374, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 2945, "ts": 1791536609.825772, "chk": "ok", "code_len": 3374}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3934, "tout": 1671, "ts": 1791536643.7713373, "chk": null, "code_len": 3374}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791536584.4636014}
{"ep": "h2-01102", "task": "L2_089", "reader": "claude-opus-4-8", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0575, "speedup": 1.0575, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 81.5, "calls": 2, "tokens_in": 14784, "tokens_out": 2545, "ctx_chars": 10527, "t_end": 1791536956.9998312, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0575, "lo95": 1.0551, "mean": 1.059, "ev_only": 1.0583, "ref_ms": 19.7331, "cand_ms": 18.65942, "n_trials": 31, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0075, "r_wall_s": 28.15, "compile_s": 2.04, "r_error": "", "src_len": 2326, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6661, "tout": 1350, "ts": 1791536888.5762782, "chk": "ok", "code_len": 2443}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8123, "tout": 1195, "ts": 1791536928.5849326, "chk": null, "code_len": 2326}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6661, "started": 1791536875.4623995}
{"ep": "h2-01003", "task": "L3_030", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.7654, "speedup": 0.7654, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 447.1, "calls": 3, "tokens_in": 48325, "tokens_out": 15769, "ctx_chars": 0, "t_end": 1791536958.2885268, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 0.7654, "lo95": 0.7624, "mean": 0.7821, "ev_only": 0.7646, "ref_ms": 19.36385, "cand_ms": 25.29882, "n_trials": 36, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0573, "r_wall_s": 311.1, "compile_s": 22.36, "r_error": "", "src_len": 17994, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10632, "tout": 8064, "ts": 1791536577.0333674, "chk": "ok", "code_len": 17994}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18808, "tout": 53, "ts": 1791536587.894944, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 18885, "tout": 7652, "ts": 1791536647.142792, "chk": null, "code_len": 17994}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 10632, "started": 1791536511.1728966}
{"ep": "h2-01077", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0598, "speedup": 1.0598, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 195.7, "calls": 2, "tokens_in": 6330, "tokens_out": 6438, "ctx_chars": 0, "t_end": 1791536959.7231889, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.0598, "lo95": 1.0577, "mean": 1.0593, "ev_only": 1.0618, "ref_ms": 8.57439, "cand_ms": 8.0907, "n_trials": 179, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0069, "r_wall_s": 130.52, "compile_s": 20.22, "r_error": "", "src_len": 5090, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 3905, "ts": 1791536794.7744162, "chk": "ok", "code_len": 5090}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4695, "tout": 2533, "ts": 1791536829.1267538, "chk": null, "code_len": 5090}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791536764.035285}
{"ep": "h2-01101", "task": "L2_028", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0236, "speedup": 1.0236, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 99.1, "calls": 2, "tokens_in": 4647, "tokens_out": 2364, "ctx_chars": 0, "t_end": 1791536963.9206383, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0236, "lo95": 1.0216, "mean": 1.0217, "ev_only": 1.0216, "ref_ms": 8.1787, "cand_ms": 7.99027, "n_trials": 149, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0048, "r_wall_s": 38.41, "compile_s": 1.93, "r_error": "", "src_len": 1299, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1752, "tout": 1614, "ts": 1791536879.5403333, "chk": "ok", "code_len": 1299}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2895, "tout": 750, "ts": 1791536925.1956968, "chk": null, "code_len": 1299}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1752, "started": 1791536864.8629494}
{"ep": "h2-01040", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0283, "speedup": 1.0283, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 332.8, "calls": 1, "tokens_in": 1666, "tokens_out": 565, "ctx_chars": 0, "t_end": 1791536963.9539266, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0283, "lo95": 1.0273, "mean": 1.0282, "ev_only": 1.0279, "ref_ms": 24.83348, "cand_ms": 24.15033, "n_trials": 57, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0286, "r_wall_s": 325.66, "compile_s": 9.0, "r_error": "", "src_len": 1012, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1666, "tout": 565, "ts": 1791536637.1501217, "chk": null, "code_len": 1012}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1666, "started": 1791536631.1137834}
{"ep": "h2-01098", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0165, "speedup": 1.0165, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 123.5, "calls": 2, "tokens_in": 16228, "tokens_out": 2508, "ctx_chars": 12117, "t_end": 1791536968.6750996, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 1.0165, "lo95": 1.0152, "mean": 1.0169, "ev_only": 1.0163, "ref_ms": 42.69267, "cand_ms": 42.00008, "n_trials": 27, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 72.39, "compile_s": 2.87, "r_error": "", "src_len": 1883, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7399, "tout": 1529, "ts": 1791536874.6793108, "chk": "ok", "code_len": 1883}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8829, "tout": 979, "ts": 1791536895.4890618, "chk": null, "code_len": 1883}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7399, "started": 1791536845.182066}
{"ep": "h2-01036", "task": "L2_039", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0026, "speedup": 1.0026, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 341.8, "calls": 2, "tokens_in": 15270, "tokens_out": 6113, "ctx_chars": 10823, "t_end": 1791536969.934022, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.0026, "lo95": 1.0018, "mean": 1.0017, "ev_only": 1.0029, "ref_ms": 30.06806, "cand_ms": 29.98886, "n_trials": 52, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0225, "r_wall_s": 40.33, "compile_s": 2.12, "r_error": "", "src_len": 3781, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6355, "tout": 4202, "ts": 1791536665.6430974, "chk": "ok", "code_len": 3781}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8915, "tout": 1911, "ts": 1791536928.9818485, "chk": null, "code_len": 3781}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6355, "started": 1791536628.1367884}
{"ep": "h2-01118", "task": "L2_076", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.9981, "speedup": 0.9981, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 57.2, "calls": 2, "tokens_in": 13358, "tokens_out": 981, "ctx_chars": 10847, "t_end": 1791536971.080351, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.9981, "lo95": 0.9967, "mean": 0.9978, "ev_only": 1.0015, "ref_ms": 7.96677, "cand_ms": 7.98231, "n_trials": 60, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0044, "r_wall_s": 38.72, "compile_s": 1.98, "r_error": "", "src_len": 821, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6369, "tout": 511, "ts": 1791536920.3728166, "chk": "ok", "code_len": 821}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6989, "tout": 470, "ts": 1791536932.2549572, "chk": null, "code_len": 821}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 6369, "started": 1791536914.489819}
{"ep": "h2-01073", "task": "L3_014", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.4671, "speedup": 1.4671, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 217.9, "calls": 2, "tokens_in": 11279, "tokens_out": 4931, "ctx_chars": 6513, "t_end": 1791536973.2859595, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-20", "gpu": 0, "r_speedup": 1.4671, "lo95": 1.4658, "mean": 1.4648, "ev_only": 1.468, "ref_ms": 15.56572, "cand_ms": 10.6098, "n_trials": 90, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1657, "r_wall_s": 121.04, "compile_s": 3.23, "r_error": "", "src_len": 3529, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4607, "tout": 3341, "ts": 1791536785.371848, "chk": "ok", "code_len": 3529}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6672, "tout": 1590, "ts": 1791536851.4646046, "chk": null, "code_len": 3529}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4607, "started": 1791536755.3445168}
{"ep": "h2-01104", "task": "L2_092", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.8232, "speedup": 2.8232, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 92.8, "calls": 2, "tokens_in": 5818, "tokens_out": 5740, "ctx_chars": 0, "t_end": 1791536973.7925117, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.8232, "lo95": 2.8189, "mean": 2.8166, "ev_only": 2.8185, "ref_ms": 8.505, "cand_ms": 3.01249, "n_trials": 158, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.2739, "r_wall_s": 35.78, "compile_s": 2.04, "r_error": "", "src_len": 2878, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1820, "tout": 4150, "ts": 1791536918.1875575, "chk": "ok", "code_len": 2878}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3998, "tout": 1590, "ts": 1791536937.7486963, "chk": null, "code_len": 2878}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1820, "started": 1791536880.976561}
{"ep": "h2-00997", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 6.6846, "speedup": 6.6846, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 493.9, "calls": 3, "tokens_in": 18624, "tokens_out": 20578, "ctx_chars": 0, "t_end": 1791536975.3885086, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 6.6846, "lo95": 6.5841, "mean": 7.3963, "ev_only": 6.7129, "ref_ms": 32.33302, "cand_ms": 4.83692, "n_trials": 42, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9996, "r_wall_s": 54.12, "compile_s": 12.1, "r_error": "", "src_len": 7546, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 12558, "ts": 1791536593.3078454, "chk": "no_custom_kernel", "code_len": 7571}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6319, "tout": 4241, "ts": 1791536638.9929245, "chk": "ok", "code_len": 7546}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10339, "tout": 3779, "ts": 1791536921.0137732, "chk": null, "code_len": 7546}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791536481.5081804}
{"ep": "h2-01119", "task": "L2_047", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "runtime_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 61.7, "calls": 1, "tokens_in": 1666, "tokens_out": 553, "ctx_chars": 0, "t_end": 1791536975.8424635, "error": "CompilationError: at 7:15:\ndef _mish_tanh_kernel(x_ptr, out_ptr, n, BLOCK: tl.constexpr):\n    pid = tl.program_id(0)\n    offs = pid * BLOCK + tl.arange(0, BLOCK", "f_status": "runtime_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "triton", "own_frac": null, "r_wall_s": 39.03, "compile_s": 2.44, "r_error": "CompilationError: at 7:15:\ndef _mish_tanh_kernel(x_ptr, out_ptr, n, BLOCK: tl.constexpr):\n    pid = tl.program_id(0)\n    offs = pid * BLOCK + tl.arange(0, BLOCK", "src_len": 988, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1666, "tout": 553, "ts": 1791536936.6104589, "chk": null, "code_len": 988}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1666, "started": 1791536914.119903}
{"ep": "h2-01111", "task": "L2_092", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 2.3719, "speedup": 2.3719, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 85.1, "calls": 2, "tokens_in": 14295, "tokens_out": 4955, "ctx_chars": 8529, "t_end": 1791536977.3252966, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.3719, "lo95": 2.3709, "mean": 2.3703, "ev_only": 2.3796, "ref_ms": 8.49352, "cand_ms": 3.58086, "n_trials": 195, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.415, "r_wall_s": 17.73, "compile_s": 2.48, "r_error": "", "src_len": 3435, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5495, "tout": 3193, "ts": 1791536919.810504, "chk": "ok", "code_len": 3435}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8800, "tout": 1762, "ts": 1791536959.543527, "chk": null, "code_len": 3435}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5495, "started": 1791536892.274888}
{"ep": "h2-01116", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.525, "speedup": 0.525, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 67.7, "calls": 1, "tokens_in": 1565, "tokens_out": 1387, "ctx_chars": 0, "t_end": 1791536978.4323416, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.525, "lo95": 0.5246, "mean": 0.5251, "ev_only": 0.525, "ref_ms": 15.48337, "cand_ms": 29.49137, "n_trials": 45, "fast1": false, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 45.4, "compile_s": 1.96, "r_error": "", "src_len": 2405, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1565, "tout": 1387, "ts": 1791536932.6360323, "chk": null, "code_len": 2405}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1565, "started": 1791536910.705683}
{"ep": "h2-01057", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.8699, "speedup": 2.8699, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 282.9, "calls": 3, "tokens_in": 17110, "tokens_out": 15461, "ctx_chars": 0, "t_end": 1791536979.4456935, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.8699, "lo95": 2.833, "mean": 2.9927, "ev_only": 2.8689, "ref_ms": 60.48985, "cand_ms": 21.07749, "n_trials": 23, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8843, "r_wall_s": 29.66, "compile_s": 14.51, "r_error": "", "src_len": 6674, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 8701, "ts": 1791536771.955364, "chk": "no_custom_kernel", "code_len": 6754}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5869, "tout": 3465, "ts": 1791536826.394572, "chk": "ok", "code_len": 6674}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 9255, "tout": 3295, "ts": 1791536949.7290106, "chk": null, "code_len": 6674}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791536696.511915}
{"ep": "h2-01107", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "compile_error", "score": 0.0, "speedup": 0.0, "correct": false, "infra": false, "infra_reruns": 0, "wall_s": 94.4, "calls": 1, "tokens_in": 1501, "tokens_out": 864, "ctx_chars": 0, "t_end": 1791536983.0353336, "error": "RuntimeError: Error building extension 'dmish_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=dmish_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "f_status": "compile_error", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.0, "lo95": null, "mean": null, "ev_only": null, "ref_ms": null, "cand_ms": null, "n_trials": null, "fast1": false, "flags": null, "backend": "cuda", "own_frac": null, "r_wall_s": 51.11, "compile_s": 51.11, "r_error": "RuntimeError: Error building extension 'dmish_ext': [1/3] c++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=dmish_ext -DTORCH_API_INCLUDE_EXTENSION_H -I/usr/local/eval", "src_len": 1742, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1501, "tout": 864, "ts": 1791536931.8775556, "chk": null, "code_len": 1742}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1501, "started": 1791536888.618364}
{"ep": "h2-01010", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.0105, "speedup": 2.0105, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 462.5, "calls": 3, "tokens_in": 16784, "tokens_out": 12684, "ctx_chars": 0, "t_end": 1791536985.1966498, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 2.0105, "lo95": 2.0091, "mean": 2.0104, "ev_only": 2.0088, "ref_ms": 14.25419, "cand_ms": 7.08979, "n_trials": 87, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 43.13, "compile_s": 11.91, "r_error": "", "src_len": 6155, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6391, "ts": 1791536597.5913913, "chk": "no_custom_kernel", "code_len": 6180}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5727, "tout": 3341, "ts": 1791536642.2921233, "chk": "ok", "code_len": 6155}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8874, "tout": 2952, "ts": 1791536941.951738, "chk": null, "code_len": 6155}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2183, "started": 1791536522.700397}
{"ep": "h2-01108", "task": "L2_006", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.1589, "speedup": 1.1589, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 99.6, "calls": 1, "tokens_in": 1757, "tokens_out": 4021, "ctx_chars": 0, "t_end": 1791536990.0108202, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 0, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.1589, "lo95": 1.1578, "mean": 1.2136, "ev_only": 1.1631, "ref_ms": 1.80257, "cand_ms": 1.55543, "n_trials": 225, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.1731, "r_wall_s": 61.46, "compile_s": 11.78, "r_error": "", "src_len": 2846, "atts": [{"turn": 0, "action": "SUBMIT", "stop": "end_turn", "tin": 1757, "tout": 4021, "ts": 1791536928.5079312, "chk": null, "code_len": 2846}], "n_turns": 1, "n_check": 0, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": [null], "first_tin": 1757, "started": 1791536890.3658807}
{"ep": "h2-01097", "task": "L1_012", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.9422, "speedup": 0.9422, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 150.1, "calls": 2, "tokens_in": 4643, "tokens_out": 2334, "ctx_chars": 0, "t_end": 1791536990.7926524, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 0.9422, "lo95": 0.8489, "mean": 0.955, "ev_only": 0.9804, "ref_ms": 0.17117, "cand_ms": 0.18167, "n_trials": 195, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 82.36, "compile_s": 14.03, "r_error": "", "src_len": 2090, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1649, "tout": 1236, "ts": 1791536851.2225916, "chk": "ok", "code_len": 2090}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 2994, "tout": 1098, "ts": 1791536908.1868002, "chk": null, "code_len": 2090}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1649, "started": 1791536840.6796942}
{"ep": "h2-01115", "task": "L3_041", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 0.8181, "speedup": 0.8181, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 87.4, "calls": 2, "tokens_in": 12066, "tokens_out": 1489, "ctx_chars": 8224, "t_end": 1791536991.9875228, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8181, "lo95": 0.8146, "mean": 0.8039, "ev_only": 0.818, "ref_ms": 61.51939, "cand_ms": 75.1976, "n_trials": 13, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 45.82, "compile_s": 2.38, "r_error": "", "src_len": 1083, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5528, "tout": 901, "ts": 1791536929.9989064, "chk": "ok", "code_len": 1083}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6538, "tout": 588, "ts": 1791536946.1050367, "chk": null, "code_len": 1083}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5528, "started": 1791536904.6132326}
{"ep": "h2-00995", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.4074, "speedup": 1.4074, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 522.2, "calls": 2, "tokens_in": 28940, "tokens_out": 18181, "ctx_chars": 0, "t_end": 1791536994.1169698, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.4074, "lo95": 1.3952, "mean": 1.717, "ev_only": 1.4109, "ref_ms": 19.3632, "cand_ms": 13.75765, "n_trials": 73, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1317, "r_wall_s": 363.89, "compile_s": 5.02, "r_error": "", "src_len": 16112, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 10994, "ts": 1791536558.3231692, "chk": "ok", "code_len": 16112}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18306, "tout": 7187, "ts": 1791536630.188064, "chk": null, "code_len": 16112}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791536471.9160283}
{"ep": "h2-01001", "task": "L2_077", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5403, "speedup": 1.5403, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 494.6, "calls": 2, "tokens_in": 6252, "tokens_out": 8931, "ctx_chars": 0, "t_end": 1791536996.5619924, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.5403, "lo95": 1.5383, "mean": 1.5401, "ev_only": 1.5403, "ref_ms": 64.31688, "cand_ms": 41.75591, "n_trials": 20, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9822, "r_wall_s": 354.1, "compile_s": 18.57, "r_error": "", "src_len": 4180, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1746, "tout": 6722, "ts": 1791536563.753507, "chk": "ok", "code_len": 4180}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4506, "tout": 2209, "ts": 1791536642.405265, "chk": null, "code_len": 4180}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1746, "started": 1791536502.0115387}
{"ep": "h2-01016", "task": "L1_008", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9971, "speedup": 0.9971, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 455.4, "calls": 4, "tokens_in": 25866, "tokens_out": 1601, "ctx_chars": 9033, "t_end": 1791537000.3579988, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9971, "lo95": 0.9951, "mean": 0.9959, "ev_only": 0.9955, "ref_ms": 15.52802, "cand_ms": 15.57335, "n_trials": 77, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0001, "r_wall_s": 49.09, "compile_s": 2.07, "r_error": "", "src_len": 526, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5488, "tout": 748, "ts": 1791536571.043906, "chk": "ok", "code_len": 721}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6345, "tout": 523, "ts": 1791536626.397411, "chk": "ok", "code_len": 526}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6977, "tout": 55, "ts": 1791536947.7576427, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 7056, "tout": 275, "ts": 1791536950.9340603, "chk": null, "code_len": 526}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "ok", "no_code", null], "first_tin": 5488, "started": 1791536545.5021503}
{"ep": "h2-01070", "task": "L2_048", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.2243, "speedup": 4.2243, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 254.8, "calls": 3, "tokens_in": 14983, "tokens_out": 11055, "ctx_chars": 0, "t_end": 1791537006.4538488, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-06", "gpu": 0, "r_speedup": 4.2243, "lo95": 4.2216, "mean": 4.215, "ev_only": 4.2176, "ref_ms": 7.80422, "cand_ms": 1.84746, "n_trials": 179, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 96.7, "compile_s": 14.59, "r_error": "", "src_len": 5752, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1725, "tout": 5015, "ts": 1791536794.3942096, "chk": "no_custom_kernel", "code_len": 5774}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5100, "tout": 3141, "ts": 1791536848.9231715, "chk": "ok", "code_len": 5752}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8158, "tout": 2899, "ts": 1791536908.9967468, "chk": null, "code_len": 5752}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1725, "started": 1791536751.661316}
{"ep": "h2-01086", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.677, "speedup": 1.677, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 220.4, "calls": 2, "tokens_in": 29288, "tokens_out": 19569, "ctx_chars": 0, "t_end": 1791537009.2139683, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.677, "lo95": 1.6534, "mean": 1.68, "ev_only": 1.6774, "ref_ms": 18.91131, "cand_ms": 11.27697, "n_trials": 71, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1744, "r_wall_s": 37.14, "compile_s": 2.13, "r_error": "", "src_len": 16743, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 10634, "tout": 12025, "ts": 1791536888.7395322, "chk": "ok", "code_len": 16893}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 18654, "tout": 7544, "ts": 1791536971.7045517, "chk": null, "code_len": 16743}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 10634, "started": 1791536788.8574462}
{"ep": "h2-00904", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.5816, "speedup": 1.5816, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 831.4, "calls": 3, "tokens_in": 33329, "tokens_out": 17255, "ctx_chars": 11539, "t_end": 1791537021.2172875, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.5816, "lo95": 1.5811, "mean": 1.5802, "ev_only": 1.5793, "ref_ms": 7.75574, "cand_ms": 4.90374, "n_trials": 176, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 368.85, "compile_s": 40.87, "r_error": "", "src_len": 7193, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7115, "tout": 9738, "ts": 1791536284.7332523, "chk": "no_custom_kernel", "code_len": 7213}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 11226, "tout": 4001, "ts": 1791536388.5936432, "chk": "ok", "code_len": 7193}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 14988, "tout": 3516, "ts": 1791536651.6696758, "chk": null, "code_len": 7193}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 7115, "started": 1791536189.8131669}
{"ep": "h2-01043", "task": "L1_058", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.8426, "speedup": 0.8426, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 389.3, "calls": 2, "tokens_in": 15411, "tokens_out": 2433, "ctx_chars": 11175, "t_end": 1791537025.7263796, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.8426, "lo95": 0.8417, "mean": 0.842, "ev_only": 0.8428, "ref_ms": 14.29053, "cand_ms": 16.95955, "n_trials": 75, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0002, "r_wall_s": 36.59, "compile_s": 1.94, "r_error": "", "src_len": 2447, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7030, "tout": 1239, "ts": 1791536647.8513181, "chk": "ok", "code_len": 2447}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8381, "tout": 1194, "ts": 1791536988.9956355, "chk": null, "code_len": 2447}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 7030, "started": 1791536636.4698071}
{"ep": "h2-01090", "task": "L1_058", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.411, "speedup": 2.411, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 213.3, "calls": 2, "tokens_in": 7817, "tokens_out": 9831, "ctx_chars": 0, "t_end": 1791537028.1402931, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.411, "lo95": 2.4055, "mean": 2.4119, "ev_only": 2.4237, "ref_ms": 14.348, "cand_ms": 5.95117, "n_trials": 113, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 54.68, "compile_s": 13.19, "r_error": "", "src_len": 5733, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2183, "tout": 6941, "ts": 1791536876.336049, "chk": "ok", "code_len": 5733}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5634, "tout": 2890, "ts": 1791536973.0305252, "chk": null, "code_len": 5733}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2183, "started": 1791536814.8653095}
{"ep": "h2-00954", "task": "L2_094", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.0404, "speedup": 1.0404, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 674.2, "calls": 2, "tokens_in": 5104, "tokens_out": 3149, "ctx_chars": 0, "t_end": 1791537032.9711328, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0404, "lo95": 1.0397, "mean": 1.0382, "ev_only": 1.0432, "ref_ms": 8.28986, "cand_ms": 7.968, "n_trials": 155, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0053, "r_wall_s": 400.7, "compile_s": 6.88, "r_error": "", "src_len": 1867, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1766, "tout": 2151, "ts": 1791536393.9904795, "chk": "ok", "code_len": 1867}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3338, "tout": 998, "ts": 1791536632.1920466, "chk": null, "code_len": 1867}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1766, "started": 1791536358.723972}
{"ep": "h2-01110", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.0261, "speedup": 1.0261, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 142.2, "calls": 2, "tokens_in": 7206, "tokens_out": 9196, "ctx_chars": 0, "t_end": 1791537033.5864072, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.0261, "lo95": 1.0254, "mean": 1.0252, "ev_only": 1.0259, "ref_ms": 30.44203, "cand_ms": 29.66698, "n_trials": 44, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.0164, "r_wall_s": 50.55, "compile_s": 12.32, "r_error": "", "src_len": 7271, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1625, "tout": 5725, "ts": 1791536937.7694666, "chk": "ok", "code_len": 7325}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5581, "tout": 3471, "ts": 1791536982.5277405, "chk": null, "code_len": 7271}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1625, "started": 1791536891.3706305}
{"ep": "h2-01094", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 4.2377, "speedup": 4.2377, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 206.6, "calls": 3, "tokens_in": 13095, "tokens_out": 8913, "ctx_chars": 0, "t_end": 1791537037.4512165, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 4.2377, "lo95": 4.2135, "mean": 4.3343, "ev_only": 4.331, "ref_ms": 1.83488, "cand_ms": 0.43299, "n_trials": 170, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9873, "r_wall_s": 75.32, "compile_s": 14.73, "r_error": "", "src_len": 4343, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 4187, "ts": 1791536866.7031524, "chk": "no_custom_kernel", "code_len": 4365}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4467, "tout": 2489, "ts": 1791536903.1831794, "chk": "ok", "code_len": 4343}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 6869, "tout": 2237, "ts": 1791536962.0881119, "chk": null, "code_len": 4343}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1759, "started": 1791536830.8799927}
{"ep": "h2-01037", "task": "L2_029", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9979, "speedup": 0.9979, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 410.7, "calls": 2, "tokens_in": 11396, "tokens_out": 1101, "ctx_chars": 9028, "t_end": 1791537039.3640568, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9979, "lo95": 0.9967, "mean": 0.999, "ev_only": 1.0009, "ref_ms": 7.99597, "cand_ms": 8.01278, "n_trials": 82, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0041, "r_wall_s": 59.32, "compile_s": 1.92, "r_error": "", "src_len": 1037, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5369, "tout": 546, "ts": 1791536633.7855082, "chk": "ok", "code_len": 1024}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 6027, "tout": 555, "ts": 1791536979.9606504, "chk": null, "code_len": 1037}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5369, "started": 1791536628.6771672}
{"ep": "h2-00984", "task": "L2_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.0215, "speedup": 1.0215, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 596.4, "calls": 2, "tokens_in": 13573, "tokens_out": 5316, "ctx_chars": 9266, "t_end": 1791537040.2181337, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-21", "gpu": 0, "r_speedup": 1.0215, "lo95": 1.0209, "mean": 1.0208, "ev_only": 1.0217, "ref_ms": 30.41237, "cand_ms": 29.77213, "n_trials": 46, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0209, "r_wall_s": 403.72, "compile_s": 6.73, "r_error": "", "src_len": 3585, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5616, "tout": 3481, "ts": 1791536594.546413, "chk": "ok", "code_len": 3585}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 7957, "tout": 1835, "ts": 1791536635.5154727, "chk": null, "code_len": 3585}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5616, "started": 1791536443.7930207}
{"ep": "h2-01082", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 3.0997, "speedup": 3.0997, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 261.5, "calls": 3, "tokens_in": 26756, "tokens_out": 15939, "ctx_chars": 8224, "t_end": 1791537043.8796892, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 3.0997, "lo95": 3.0902, "mean": 3.0992, "ev_only": 3.1004, "ref_ms": 60.48265, "cand_ms": 19.51266, "n_trials": 27, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8717, "r_wall_s": 24.88, "compile_s": 14.74, "r_error": "", "src_len": 6061, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 9822, "ts": 1791536884.9654863, "chk": "no_custom_kernel", "code_len": 6083}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9064, "tout": 3220, "ts": 1791536982.222058, "chk": "ok", "code_len": 6061}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 12162, "tout": 2897, "ts": 1791537018.9338145, "chk": null, "code_len": 6061}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791536782.3415554}
{"ep": "h2-01076", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.3821, "speedup": 0.3821, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 281.7, "calls": 5, "tokens_in": 55891, "tokens_out": 13426, "ctx_chars": 6513, "t_end": 1791537045.3139796, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 4, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 0.3821, "lo95": 0.3796, "mean": 0.3917, "ev_only": 0.3821, "ref_ms": 7.12006, "cand_ms": 18.63553, "n_trials": 50, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.055, "r_wall_s": 24.68, "compile_s": 3.04, "r_error": "", "src_len": 6572, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2802, "ts": 1791536787.3218968, "chk": "incorrect", "code_len": 6275}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8579, "tout": 2606, "ts": 1791536845.2493045, "chk": "incorrect", "code_len": 4788}, {"turn": 2, "action": "CHECK", "stop": "end_turn", "tin": 11317, "tout": 2229, "ts": 1791536878.1406152, "chk": "ok", "code_len": 4489}, {"turn": 3, "action": "CHECK", "stop": "end_turn", "tin": 13655, "tout": 2905, "ts": 1791536972.274512, "chk": "ok", "code_len": 6572}, {"turn": 4, "action": "SUBMIT", "stop": "end_turn", "tin": 16695, "tout": 2884, "ts": 1791537020.5897202, "chk": null, "code_len": 6572}], "n_turns": 5, "n_check": 4, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "incorrect", "ok", "ok", null], "first_tin": 5645, "started": 1791536763.6594214}
{"ep": "h2-01103", "task": "L1_061", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 2.7504, "speedup": 2.7504, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 175.2, "calls": 2, "tokens_in": 6747, "tokens_out": 7938, "ctx_chars": 0, "t_end": 1791537050.887413, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.7504, "lo95": 2.7489, "mean": 2.7491, "ev_only": 2.7506, "ref_ms": 42.56357, "cand_ms": 15.47532, "n_trials": 38, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9027, "r_wall_s": 77.15, "compile_s": 2.08, "r_error": "", "src_len": 4082, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2036, "tout": 5887, "ts": 1791536935.1965287, "chk": "ok", "code_len": 4082}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4711, "tout": 2051, "ts": 1791536972.8776557, "chk": null, "code_len": 4082}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 2036, "started": 1791536875.652506}
{"ep": "h2-01063", "task": "L3_037", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 5.917, "speedup": 5.917, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 344.3, "calls": 3, "tokens_in": 18400, "tokens_out": 23026, "ctx_chars": 0, "t_end": 1791537060.578349, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 5.917, "lo95": 5.8461, "mean": 6.5886, "ev_only": 5.9183, "ref_ms": 32.44818, "cand_ms": 5.48389, "n_trials": 30, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.9712, "r_wall_s": 86.54, "compile_s": 12.96, "r_error": "", "src_len": 7651, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1966, "tout": 15247, "ts": 1791536852.197248, "chk": "no_custom_kernel", "code_len": 7656}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 6308, "tout": 4093, "ts": 1791536909.005372, "chk": "ok", "code_len": 7651}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10126, "tout": 3686, "ts": 1791536973.7843802, "chk": null, "code_len": 7651}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1966, "started": 1791536716.3281896}
{"ep": "h2-01045", "task": "L1_008", "reader": "claude-opus-5-5", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 1.4076, "speedup": 1.4076, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 424.1, "calls": 2, "tokens_in": 5107, "tokens_out": 4537, "ctx_chars": 0, "t_end": 1791537062.3924298, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.4076, "lo95": 1.4048, "mean": 1.4031, "ev_only": 1.4084, "ref_ms": 15.52618, "cand_ms": 11.0303, "n_trials": 88, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.9654, "r_wall_s": 97.11, "compile_s": 3.34, "r_error": "", "src_len": 2401, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1567, "tout": 3087, "ts": 1791536666.140009, "chk": "ok", "code_len": 2401}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3540, "tout": 1450, "ts": 1791536964.8370779, "chk": null, "code_len": 2401}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1567, "started": 1791536638.322904}
{"ep": "h2-01100", "task": "L3_009", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 0.9452, "speedup": 0.9452, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 209.4, "calls": 2, "tokens_in": 13905, "tokens_out": 7867, "ctx_chars": 6513, "t_end": 1791537063.2275295, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9452, "lo95": 0.9367, "mean": 0.8932, "ev_only": 0.949, "ref_ms": 3.28318, "cand_ms": 3.4736, "n_trials": 213, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0809, "r_wall_s": 63.08, "compile_s": 2.39, "r_error": "", "src_len": 5533, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5448, "tout": 5272, "ts": 1791536906.7435024, "chk": "ok", "code_len": 5533}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8457, "tout": 2595, "ts": 1791536999.893398, "chk": null, "code_len": 5533}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 5448, "started": 1791536854.34382}
{"ep": "h2-01112", "task": "L2_006", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 4.0811, "speedup": 4.0811, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 166.8, "calls": 2, "tokens_in": 6902, "tokens_out": 8116, "ctx_chars": 0, "t_end": 1791537064.654067, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 4.0811, "lo95": 4.006, "mean": 4.3968, "ev_only": 4.2832, "ref_ms": 1.80747, "cand_ms": 0.44289, "n_trials": 176, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 65.15, "compile_s": 12.97, "r_error": "", "src_len": 5515, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1759, "tout": 5362, "ts": 1791536943.0095787, "chk": "ok", "code_len": 5515}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 5143, "tout": 2754, "ts": 1791536999.084171, "chk": null, "code_len": 5515}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1759, "started": 1791536897.8882377}
{"ep": "h2-01114", "task": "L1_056", "reader": "claude-opus-4-8", "arm": "predictor_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9757, "speedup": 0.9757, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 169.4, "calls": 3, "tokens_in": 24009, "tokens_out": 2178, "ctx_chars": 11539, "t_end": 1791537071.8034472, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 0.9757, "lo95": 0.9741, "mean": 0.9754, "ev_only": 0.9759, "ref_ms": 31.4213, "cand_ms": 32.20365, "n_trials": 47, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0203, "r_wall_s": 83.68, "compile_s": 2.67, "r_error": "", "src_len": 1804, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 7032, "tout": 1329, "ts": 1791536934.453631, "chk": "ok", "code_len": 1804}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8470, "tout": 13, "ts": 1791536953.3140974, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8507, "tout": 836, "ts": 1791536987.3223312, "chk": null, "code_len": 1804}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 7032, "started": 1791536902.4391682}
{"ep": "h2-01109", "task": "L3_030", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 1.7352, "speedup": 1.7352, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 188.2, "calls": 2, "tokens_in": 30851, "tokens_out": 18954, "ctx_chars": 2425, "t_end": 1791537079.0872676, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 1.7352, "lo95": 1.7239, "mean": 1.7246, "ev_only": 1.7394, "ref_ms": 18.76457, "cand_ms": 10.81433, "n_trials": 63, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.1205, "r_wall_s": 25.11, "compile_s": 1.9, "r_error": "", "src_len": 16151, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 11550, "tout": 11715, "ts": 1791536982.8727012, "chk": "ok", "code_len": 16151}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 19301, "tout": 7239, "ts": 1791537053.9112358, "chk": null, "code_len": 16151}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 11550, "started": 1791536890.8859634}
{"ep": "h2-01092", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_self_gating", "policy": "h2", "store": true, "status": "ok", "score": 2.2126, "speedup": 2.2126, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 263.6, "calls": 3, "tokens_in": 26376, "tokens_out": 14694, "ctx_chars": 8224, "t_end": 1791537081.9056973, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 2.2126, "lo95": 2.2076, "mean": 2.2074, "ev_only": 2.2129, "ref_ms": 60.00595, "cand_ms": 27.12058, "n_trials": 25, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8724, "r_wall_s": 23.26, "compile_s": 10.99, "r_error": "", "src_len": 6097, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 8603, "ts": 1791536896.1192372, "chk": "no_custom_kernel", "code_len": 6137}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8898, "tout": 3241, "ts": 1791536976.6763525, "chk": "ok", "code_len": 6097}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11948, "tout": 2850, "ts": 1791537058.5868309, "chk": null, "code_len": 6097}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791536818.3514645}
{"ep": "h2-01095", "task": "L2_043", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 1.005, "speedup": 1.005, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 250.8, "calls": 4, "tokens_in": 23982, "tokens_out": 2714, "ctx_chars": 11823, "t_end": 1791537084.626335, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.005, "lo95": 1.0032, "mean": 1.0039, "ev_only": 1.0055, "ref_ms": 27.54934, "cand_ms": 27.41186, "n_trials": 47, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0284, "r_wall_s": 88.98, "compile_s": 3.1, "r_error": "", "src_len": 2364, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6921, "tout": 1446, "ts": 1791536848.0744126, "chk": "ok", "code_len": 2492}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 8479, "tout": 79, "ts": 1791536985.7236652, "chk": "no_code", "code_len": 0}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8582, "tout": 1189, "ts": 1791536995.5893214, "chk": null, "code_len": 2364}], "n_turns": 3, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", "no_code", null], "first_tin": 6921, "started": 1791536833.8625674}
{"ep": "h2-01081", "task": "L1_066", "reader": "claude-opus-4-8", "arm": "none", "policy": "h2", "store": false, "status": "ok", "score": 0.993, "speedup": 0.993, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 308.2, "calls": 4, "tokens_in": 17901, "tokens_out": 2951, "ctx_chars": 0, "t_end": 1791537089.9444938, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.993, "lo95": 0.9922, "mean": 0.9925, "ev_only": 0.9941, "ref_ms": 7.77041, "cand_ms": 7.82542, "n_trials": 127, "fast1": false, "flags": [], "backend": "cuda", "own_frac": 0.0046, "r_wall_s": 91.07, "compile_s": 12.32, "r_error": "", "src_len": 1754, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2157, "tout": 1178, "ts": 1791536828.1146853, "chk": "compile_error", "code_len": 1695}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 4573, "tout": 834, "ts": 1791536940.042557, "chk": "ok", "code_len": 1754}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 5516, "tout": 115, "ts": 1791536990.825579, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 5655, "tout": 824, "ts": 1791536998.001981, "chk": null, "code_len": 1754}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["compile_error", "ok", "no_code", null], "first_tin": 2157, "started": 1791536781.7874918}
{"ep": "h2-01096", "task": "L2_075", "reader": "claude-opus-4-8", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 1.0083, "speedup": 1.0083, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 254.0, "calls": 3, "tokens_in": 25937, "tokens_out": 5128, "ctx_chars": 10167, "t_end": 1791537090.554972, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.0083, "lo95": 1.0073, "mean": 1.0074, "ev_only": 1.0088, "ref_ms": 8.45017, "cand_ms": 8.38052, "n_trials": 142, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.0557, "r_wall_s": 81.25, "compile_s": 2.72, "r_error": "", "src_len": 1749, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 6104, "tout": 3088, "ts": 1791536893.370617, "chk": "incorrect", "code_len": 1766}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9299, "tout": 1123, "ts": 1791536966.1518874, "chk": "ok", "code_len": 1749}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 10534, "tout": 917, "ts": 1791537009.1700747, "chk": null, "code_len": 1749}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", null], "first_tin": 6104, "started": 1791536836.562804}
{"ep": "h2-01033", "task": "L3_020", "reader": "claude-opus-4-8", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.2927, "speedup": 0.2927, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 478.8, "calls": 4, "tokens_in": 37066, "tokens_out": 8042, "ctx_chars": 6513, "t_end": 1791537092.811946, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 3, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.2927, "lo95": 0.2868, "mean": 0.3319, "ev_only": 0.2915, "ref_ms": 7.62647, "cand_ms": 26.05796, "n_trials": 46, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.0601, "r_wall_s": 25.34, "compile_s": 3.03, "r_error": "", "src_len": 5634, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5645, "tout": 2736, "ts": 1791536637.7021384, "chk": "incorrect", "code_len": 6109}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 8513, "tout": 2804, "ts": 1791537028.0856042, "chk": "ok", "code_len": 5634}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11429, "tout": 26, "ts": 1791537047.2601035, "chk": "no_code", "code_len": 0}, {"turn": 3, "action": "SUBMIT", "stop": "end_turn", "tin": 11479, "tout": 2476, "ts": 1791537067.2220182, "chk": null, "code_len": 5634}], "n_turns": 4, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["incorrect", "ok", "no_code", null], "first_tin": 5645, "started": 1791536614.0616007}
{"ep": "h2-01020", "task": "L2_075", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.5118, "speedup": 1.5118, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 520.0, "calls": 2, "tokens_in": 6068, "tokens_out": 7155, "ctx_chars": 0, "t_end": 1791537092.8415585, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 1.5118, "lo95": 1.5104, "mean": 1.5113, "ev_only": 1.509, "ref_ms": 8.48816, "cand_ms": 5.61449, "n_trials": 182, "fast1": true, "flags": [], "backend": "triton", "own_frac": 1.0, "r_wall_s": 34.21, "compile_s": 2.16, "r_error": "", "src_len": 4123, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1635, "tout": 4898, "ts": 1791536640.4387014, "chk": "ok", "code_len": 4123}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 4433, "tout": 2257, "ts": 1791537058.2350817, "chk": null, "code_len": 4123}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1635, "started": 1791536572.8700411}
{"ep": "h2-01025", "task": "L2_043", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.2982, "speedup": 2.2982, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 505.0, "calls": 2, "tokens_in": 5402, "tokens_out": 6572, "ctx_chars": 0, "t_end": 1791537100.1376066, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.2982, "lo95": 2.2944, "mean": 2.2961, "ev_only": 2.2986, "ref_ms": 27.56132, "cand_ms": 11.9928, "n_trials": 56, "fast1": true, "flags": [], "backend": "triton", "own_frac": 0.958, "r_wall_s": 81.23, "compile_s": 2.22, "r_error": "", "src_len": 2690, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1730, "tout": 5139, "ts": 1791536642.383512, "chk": "ok", "code_len": 2690}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 3672, "tout": 1433, "ts": 1791537018.6523743, "chk": null, "code_len": 2690}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 1730, "started": 1791536595.1457102}
{"ep": "h2-01117", "task": "L3_021", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": true, "status": "ok", "score": 0.9768, "speedup": 0.9768, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 191.6, "calls": 2, "tokens_in": 14138, "tokens_out": 12584, "ctx_chars": 6513, "t_end": 1791537103.4316213, "error": "ok", "f_status": "ok", "f_error": "", "submit_turn": 1, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 0.9768, "lo95": 0.9764, "mean": 0.9771, "ev_only": 0.9771, "ref_ms": 15.71057, "cand_ms": 16.08447, "n_trials": 91, "fast1": false, "flags": [], "backend": "triton", "own_frac": 0.5723, "r_wall_s": 41.41, "compile_s": 2.59, "r_error": "", "src_len": 7828, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 4864, "tout": 8806, "ts": 1791537000.2323318, "chk": "ok", "code_len": 7939}, {"turn": 1, "action": "SUBMIT", "stop": "end_turn", "tin": 9274, "tout": 3778, "ts": 1791537061.6322465, "chk": null, "code_len": 7828}], "n_turns": 2, "n_check": 1, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["ok", null], "first_tin": 4864, "started": 1791536911.8041315}
{"ep": "h2-01088", "task": "L1_066", "reader": "claude-opus-5-5", "arm": "predictor_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 1.1923, "speedup": 1.1923, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 292.6, "calls": 3, "tokens_in": 16632, "tokens_out": 12827, "ctx_chars": 0, "t_end": 1791537104.7084916, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-01", "gpu": 0, "r_speedup": 1.1923, "lo95": 1.191, "mean": 1.1911, "ev_only": 1.1933, "ref_ms": 7.7459, "cand_ms": 6.49656, "n_trials": 132, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 1.0, "r_wall_s": 101.68, "compile_s": 14.49, "r_error": "", "src_len": 5840, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 2159, "tout": 6572, "ts": 1791536870.264085, "chk": "no_custom_kernel", "code_len": 5858}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5657, "tout": 3395, "ts": 1791536950.6139274, "chk": "ok", "code_len": 5840}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8816, "tout": 2860, "ts": 1791537002.339801, "chk": null, "code_len": 5840}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 2159, "started": 1791536812.1339347}
{"ep": "h2-01106", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "opus_digest_calibrated", "policy": "h2", "store": false, "status": "ok", "score": 2.7436, "speedup": 2.7436, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 224.8, "calls": 3, "tokens_in": 16521, "tokens_out": 13989, "ctx_chars": 0, "t_end": 1791537109.7925851, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-15", "gpu": 0, "r_speedup": 2.7436, "lo95": 2.7409, "mean": 2.7446, "ev_only": 2.7461, "ref_ms": 59.41387, "cand_ms": 21.65581, "n_trials": 24, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8873, "r_wall_s": 38.67, "compile_s": 15.37, "r_error": "", "src_len": 6450, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 1986, "tout": 7619, "ts": 1791536951.4908857, "chk": "no_custom_kernel", "code_len": 6462}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 5620, "tout": 3295, "ts": 1791537019.5625288, "chk": "ok", "code_len": 6450}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 8915, "tout": 3075, "ts": 1791537070.8572662, "chk": null, "code_len": 6450}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 1986, "started": 1791536885.0018091}
{"ep": "h2-01113", "task": "L3_041", "reader": "claude-opus-5-5", "arm": "show_all", "policy": "h2", "store": true, "status": "ok", "score": 3.572, "speedup": 3.572, "correct": true, "infra": false, "infra_reruns": 0, "wall_s": 252.2, "calls": 3, "tokens_in": 26518, "tokens_out": 15846, "ctx_chars": 8224, "t_end": 1791537152.711742, "error": "", "f_status": "ok", "f_error": "", "submit_turn": 2, "resumed": false, "endpoint": "endpoint-03", "gpu": 0, "r_speedup": 3.572, "lo95": 3.5653, "mean": 3.5697, "ev_only": 3.5741, "ref_ms": 59.68155, "cand_ms": 16.70837, "n_trials": 24, "fast1": true, "flags": [], "backend": "cuda", "own_frac": 0.8492, "r_wall_s": 19.71, "compile_s": 10.4, "r_error": "", "src_len": 5746, "atts": [{"turn": 0, "action": "CHECK", "stop": "end_turn", "tin": 5530, "tout": 9431, "ts": 1791537001.0038383, "chk": "no_custom_kernel", "code_len": 5768}, {"turn": 1, "action": "CHECK", "stop": "end_turn", "tin": 9002, "tout": 3574, "ts": 1791537098.411083, "chk": "ok", "code_len": 5746}, {"turn": 2, "action": "SUBMIT", "stop": "end_turn", "tin": 11986, "tout": 2841, "ts": 1791537132.9391751, "chk": null, "code_len": 5746}], "n_turns": 3, "n_check": 2, "max_tok_hits": 0, "stops": ["end_turn"], "chk_statuses": ["no_custom_kernel", "ok", null], "first_tin": 5530, "started": 1791536900.532792}