1{2 "system": "p4v2_gr_s3",3 "split": "test_iid",4 "checkpoint": "experiments/runs/p4v2_gr_seed03_20261030001749/checkpoint",5 "checkpoint_sha256": "29c37958b4eefbbcba2b5fb04b3f1ee570c53678afafc28776ec659d5cbace17",6 "config_sha256": "a9cd9e1dabd9c0a5207678b323f88b0854bc936f59ffc8a0eae25664f80c3d09",7 "code_commit": "unknown",8 "device": "mps",9 "precision": "fp32",10 "evaluated_at": "2026-10-30T04:05:47-0400",11 "data_sha256": "06620e1cf60b7250aac71f4fb5ee74e3705f4b9aab102e22b5c5bdbed76c4ae2",12 "exact_match": {13 "budget0": {14 "accuracy": 0.368,15 "n": 1000,16 "by_family": {17 "bind": 0.2119815668202765,18 "fsm": 0.4358974358974359,19 "arith": 0.06701030927835051,20 "excl": 0.3348623853211009,21 "sched": 0.781395348837209322 },23 "edits_per_token": 0.024 },25 "forced_b2": {26 "accuracy": 0.38,27 "n": 1000,28 "by_family": {29 "bind": 0.22119815668202766,30 "fsm": 0.4807692307692308,31 "arith": 0.07731958762886598,32 "excl": 0.3211009174311927,33 "sched": 0.834 },35 "edits_per_token": 2.036 }37 },38 "seconds": 14.939}