FlowReasoningModel / sudoku /source_evaluation.json
helblazer811's picture
Add selected FRM checkpoints and portable calibrated evaluations
3e8c45f verified
Raw History Blame Contribute Delete
31.9 kB
{
"schema_version": 1,
"git_commit": null,
"checkpoint": "/proj/grp-flow-map/FlowMapReasoning/public_simplified_repro/sudoku_seed0/stage_b/final.pt",
"checkpoint_sha256": "00172c543dc43abe023f2b3058bf62785d200164eaf6837d21698e4b92e2113a",
"benchmark": "/proj/grp-flow-map/LoopedFlows/experiments/public_simplified_repro/outputs/evaluations/sudoku_lite_adaptive_20261016_v1/data/lite/benchmark.npz",
"benchmark_sha256": "183017b414e8a7ea4efe3d17c6bb26a64da801d8443f9da27e7e3f403e846136",
"policy_config": {
"max_nfe": 4096,
"max_breadth": 128,
"held_t": 0.4,
"renoise": 1.0,
"check_every": 16,
"residual_window": 64,
"residual_metric": "euclidean",
"confirmation_checks": 8,
"require_prediction_stability": true,
"min_nfe": 64,
"batch_size": 128,
"seed": 0,
"restart_seed_offset": 1000003,
"flops_per_forward": 896039936
},
"residual": "trailing-window mean of per-puzzle euclidean distance between consecutive denoiser distributions, averaged over non-clue cells",
"oracle_used_for_test_selection": false,
"kind": "residual_guided_inference_evaluation",
"run_fingerprint": "7af0b062ad8b4b5b80bebac5c69c570c521f33033bc36c470ae6432698549ac0",
"num_puzzles": 2048,
"max_samples_override": null,
"model_step": 175000,
"calibration": "/proj/grp-flow-map/LoopedFlows/experiments/public_simplified_repro/outputs/evaluations/sudoku_lite_adaptive_20261016_v1/calibration/threshold.json",
"calibration_sha256": "e679275da4b239efa1127e028969e36392a685f7c00ec7604412d232dede3e84",
"calibration_benchmark_sha256": "e1f3b99cd5ce91dfe1e7de44995e1e38bbf1372766457025eca589625e89281e",
"threshold": 0.4277484118938446,
"threshold_selection": {
"threshold": 0.4277484118938446,
"accuracy": 0.9931640625,
"solved": 1017,
"mean_nfe": 327.265625,
"accepted_fraction": 0.994140625,
"fixed_accuracy": 0.9931640625,
"max_accuracy_drop": 0.001,
"selection_objective": "minimize mean NFE subject to accuracy >= fixed-depth accuracy - max_accuracy_drop; tie-break by higher accuracy then lower threshold"
},
"policies": [
{
"policy": "Fixed-depth FRM",
"solved": 2011,
"num_puzzles": 2048,
"solve_rate": 0.98193359375,
"mean_nfe": 4096.0,
"mean_tflops": 3.670179577856
},
{
"policy": "+ Residual early stopping",
"solved": 2009,
"num_puzzles": 2048,
"solve_rate": 0.98095703125,
"mean_nfe": 378.6796875,
"mean_tflops": 0.339312122952
},
{
"policy": "+ Sequential restarts",
"solved": 2042,
"num_puzzles": 2048,
"solve_rate": 0.9970703125,
"mean_nfe": 899.75,
"mean_tflops": 0.806211932416
}
],
"accepted_by_threshold": 2048,
"unresolved_at_breadth_cap": 0,
"attempts": [
{
"attempt": 1,
"attempted": 2048,
"accepted": 2013,
"remaining": 35,
"mean_nfe_among_attempted": 378.6796875
},
{
"attempt": 2,
"seed": 1000003,
"attempted": 35,
"accepted": 16,
"remaining": 19,
"mean_nfe_among_attempted": 2809.6
},
{
"attempt": 3,
"seed": 2000006,
"attempted": 19,
"accepted": 9,
"remaining": 10,
"mean_nfe_among_attempted": 2718.315789473684
},
{
"attempt": 4,
"seed": 3000009,
"attempted": 10,
"accepted": 2,
"remaining": 8,
"mean_nfe_among_attempted": 3614.4
},
{
"attempt": 5,
"seed": 4000012,
"attempted": 8,
"accepted": 0,
"remaining": 8,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 6,
"seed": 5000015,
"attempted": 8,
"accepted": 1,
"remaining": 7,
"mean_nfe_among_attempted": 4022.0
},
{
"attempt": 7,
"seed": 6000018,
"attempted": 7,
"accepted": 0,
"remaining": 7,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 8,
"seed": 7000021,
"attempted": 7,
"accepted": 0,
"remaining": 7,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 9,
"seed": 8000024,
"attempted": 7,
"accepted": 1,
"remaining": 6,
"mean_nfe_among_attempted": 3588.5714285714284
},
{
"attempt": 10,
"seed": 9000027,
"attempted": 6,
"accepted": 0,
"remaining": 6,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 11,
"seed": 10000030,
"attempted": 6,
"accepted": 0,
"remaining": 6,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 12,
"seed": 11000033,
"attempted": 6,
"accepted": 1,
"remaining": 5,
"mean_nfe_among_attempted": 3842.6666666666665
},
{
"attempt": 13,
"seed": 12000036,
"attempted": 5,
"accepted": 0,
"remaining": 5,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 14,
"seed": 13000039,
"attempted": 5,
"accepted": 0,
"remaining": 5,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 15,
"seed": 14000042,
"attempted": 5,
"accepted": 0,
"remaining": 5,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 16,
"seed": 15000045,
"attempted": 5,
"accepted": 0,
"remaining": 5,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 17,
"seed": 16000048,
"attempted": 5,
"accepted": 1,
"remaining": 4,
"mean_nfe_among_attempted": 3312.0
},
{
"attempt": 18,
"seed": 17000051,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 19,
"seed": 18000054,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 20,
"seed": 19000057,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 21,
"seed": 20000060,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 22,
"seed": 21000063,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 23,
"seed": 22000066,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 24,
"seed": 23000069,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 25,
"seed": 24000072,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 26,
"seed": 25000075,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 27,
"seed": 26000078,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 28,
"seed": 27000081,
"attempted": 4,
"accepted": 0,
"remaining": 4,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 29,
"seed": 28000084,
"attempted": 4,
"accepted": 1,
"remaining": 3,
"mean_nfe_among_attempted": 3832.0
},
{
"attempt": 30,
"seed": 29000087,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 31,
"seed": 30000090,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 32,
"seed": 31000093,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 33,
"seed": 32000096,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 34,
"seed": 33000099,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 35,
"seed": 34000102,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 36,
"seed": 35000105,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 37,
"seed": 36000108,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 38,
"seed": 37000111,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 39,
"seed": 38000114,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 40,
"seed": 39000117,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 41,
"seed": 40000120,
"attempted": 3,
"accepted": 0,
"remaining": 3,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 42,
"seed": 41000123,
"attempted": 3,
"accepted": 1,
"remaining": 2,
"mean_nfe_among_attempted": 2965.3333333333335
},
{
"attempt": 43,
"seed": 42000126,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 44,
"seed": 43000129,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 45,
"seed": 44000132,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 46,
"seed": 45000135,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 47,
"seed": 46000138,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 48,
"seed": 47000141,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 49,
"seed": 48000144,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 50,
"seed": 49000147,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 51,
"seed": 50000150,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 52,
"seed": 51000153,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 53,
"seed": 52000156,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 54,
"seed": 53000159,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 55,
"seed": 54000162,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 56,
"seed": 55000165,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 57,
"seed": 56000168,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 58,
"seed": 57000171,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 59,
"seed": 58000174,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 60,
"seed": 59000177,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 61,
"seed": 60000180,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 62,
"seed": 61000183,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 63,
"seed": 62000186,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 64,
"seed": 63000189,
"attempted": 2,
"accepted": 0,
"remaining": 2,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 65,
"seed": 64000192,
"attempted": 2,
"accepted": 1,
"remaining": 1,
"mean_nfe_among_attempted": 2256.0
},
{
"attempt": 66,
"seed": 65000195,
"attempted": 1,
"accepted": 0,
"remaining": 1,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 67,
"seed": 66000198,
"attempted": 1,
"accepted": 0,
"remaining": 1,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 68,
"seed": 67000201,
"attempted": 1,
"accepted": 0,
"remaining": 1,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 69,
"seed": 68000204,
"attempted": 1,
"accepted": 0,
"remaining": 1,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 70,
"seed": 69000207,
"attempted": 1,
"accepted": 0,
"remaining": 1,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 71,
"seed": 70000210,
"attempted": 1,
"accepted": 0,
"remaining": 1,
"mean_nfe_among_attempted": 4096.0
},
{
"attempt": 72,
"seed": 71000213,
"attempted": 1,
"accepted": 1,
"remaining": 0,
"mean_nfe_among_attempted": 1712.0
}
],
"restart_frontier": [
{
"breadth_cap": 1,
"solved": 2009,
"num_puzzles": 2048,
"solve_rate": 0.98095703125,
"mean_nfe": 378.6796875,
"mean_tflops": 0.339312122952,
"accepted": 2013,
"unresolved": 35
},
{
"breadth_cap": 2,
"solved": 2023,
"num_puzzles": 2048,
"solve_rate": 0.98779296875,
"mean_nfe": 426.6953125,
"mean_tflops": 0.382336040504,
"accepted": 2029,
"unresolved": 19
},
{
"breadth_cap": 3,
"solved": 2032,
"num_puzzles": 2048,
"solve_rate": 0.9921875,
"mean_nfe": 451.9140625,
"mean_tflops": 0.40493304764,
"accepted": 2038,
"unresolved": 10
},
{
"breadth_cap": 4,
"solved": 2034,
"num_puzzles": 2048,
"solve_rate": 0.9931640625,
"mean_nfe": 469.5625,
"mean_tflops": 0.420746752448,
"accepted": 2040,
"unresolved": 8
},
{
"breadth_cap": 5,
"solved": 2034,
"num_puzzles": 2048,
"solve_rate": 0.9931640625,
"mean_nfe": 485.5625,
"mean_tflops": 0.435083391424,
"accepted": 2040,
"unresolved": 8
},
{
"breadth_cap": 6,
"solved": 2035,
"num_puzzles": 2048,
"solve_rate": 0.99365234375,
"mean_nfe": 501.2734375,
"mean_tflops": 0.449161018856,
"accepted": 2041,
"unresolved": 7
},
{
"breadth_cap": 7,
"solved": 2035,
"num_puzzles": 2048,
"solve_rate": 0.99365234375,
"mean_nfe": 515.2734375,
"mean_tflops": 0.46170557796,
"accepted": 2041,
"unresolved": 7
},
{
"breadth_cap": 8,
"solved": 2035,
"num_puzzles": 2048,
"solve_rate": 0.99365234375,
"mean_nfe": 529.2734375,
"mean_tflops": 0.474250137064,
"accepted": 2041,
"unresolved": 7
},
{
"breadth_cap": 9,
"solved": 2036,
"num_puzzles": 2048,
"solve_rate": 0.994140625,
"mean_nfe": 541.5390625,
"mean_tflops": 0.485240626904,
"accepted": 2042,
"unresolved": 6
},
{
"breadth_cap": 10,
"solved": 2036,
"num_puzzles": 2048,
"solve_rate": 0.994140625,
"mean_nfe": 553.5390625,
"mean_tflops": 0.495993106136,
"accepted": 2042,
"unresolved": 6
},
{
"breadth_cap": 11,
"solved": 2036,
"num_puzzles": 2048,
"solve_rate": 0.994140625,
"mean_nfe": 565.5390625,
"mean_tflops": 0.506745585368,
"accepted": 2042,
"unresolved": 6
},
{
"breadth_cap": 12,
"solved": 2037,
"num_puzzles": 2048,
"solve_rate": 0.99462890625,
"mean_nfe": 576.796875,
"mean_tflops": 0.51683303496,
"accepted": 2043,
"unresolved": 5
},
{
"breadth_cap": 13,
"solved": 2037,
"num_puzzles": 2048,
"solve_rate": 0.99462890625,
"mean_nfe": 586.796875,
"mean_tflops": 0.52579343432,
"accepted": 2043,
"unresolved": 5
},
{
"breadth_cap": 14,
"solved": 2037,
"num_puzzles": 2048,
"solve_rate": 0.99462890625,
"mean_nfe": 596.796875,
"mean_tflops": 0.53475383368,
"accepted": 2043,
"unresolved": 5
},
{
"breadth_cap": 15,
"solved": 2037,
"num_puzzles": 2048,
"solve_rate": 0.99462890625,
"mean_nfe": 606.796875,
"mean_tflops": 0.54371423304,
"accepted": 2043,
"unresolved": 5
},
{
"breadth_cap": 16,
"solved": 2037,
"num_puzzles": 2048,
"solve_rate": 0.99462890625,
"mean_nfe": 616.796875,
"mean_tflops": 0.5526746324,
"accepted": 2043,
"unresolved": 5
},
{
"breadth_cap": 17,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 624.8828125,
"mean_tflops": 0.55991995532,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 18,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 632.8828125,
"mean_tflops": 0.567088274808,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 19,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 640.8828125,
"mean_tflops": 0.574256594296,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 20,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 648.8828125,
"mean_tflops": 0.581424913784,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 21,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 656.8828125,
"mean_tflops": 0.588593233272,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 22,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 664.8828125,
"mean_tflops": 0.59576155276,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 23,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 672.8828125,
"mean_tflops": 0.602929872248,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 24,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 680.8828125,
"mean_tflops": 0.610098191736,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 25,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 688.8828125,
"mean_tflops": 0.617266511224,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 26,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 696.8828125,
"mean_tflops": 0.624434830712,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 27,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 704.8828125,
"mean_tflops": 0.6316031502,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 28,
"solved": 2038,
"num_puzzles": 2048,
"solve_rate": 0.9951171875,
"mean_nfe": 712.8828125,
"mean_tflops": 0.638771469688,
"accepted": 2044,
"unresolved": 4
},
{
"breadth_cap": 29,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 720.3671875,
"mean_tflops": 0.645477768584,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 30,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 726.3671875,
"mean_tflops": 0.6508540082,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 31,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 732.3671875,
"mean_tflops": 0.656230247816,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 32,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 738.3671875,
"mean_tflops": 0.661606487432,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 33,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 744.3671875,
"mean_tflops": 0.666982727048,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 34,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 750.3671875,
"mean_tflops": 0.672358966664,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 35,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 756.3671875,
"mean_tflops": 0.67773520628,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 36,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 762.3671875,
"mean_tflops": 0.683111445896,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 37,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 768.3671875,
"mean_tflops": 0.688487685512,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 38,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 774.3671875,
"mean_tflops": 0.693863925128,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 39,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 780.3671875,
"mean_tflops": 0.699240164744,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 40,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 786.3671875,
"mean_tflops": 0.70461640436,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 41,
"solved": 2039,
"num_puzzles": 2048,
"solve_rate": 0.99560546875,
"mean_nfe": 792.3671875,
"mean_tflops": 0.709992643976,
"accepted": 2045,
"unresolved": 3
},
{
"breadth_cap": 42,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 796.7109375,
"mean_tflops": 0.713884817448,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 43,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 800.7109375,
"mean_tflops": 0.717468977192,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 44,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 804.7109375,
"mean_tflops": 0.721053136936,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 45,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 808.7109375,
"mean_tflops": 0.72463729668,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 46,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 812.7109375,
"mean_tflops": 0.728221456424,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 47,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 816.7109375,
"mean_tflops": 0.731805616168,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 48,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 820.7109375,
"mean_tflops": 0.735389775912,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 49,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 824.7109375,
"mean_tflops": 0.738973935656,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 50,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 828.7109375,
"mean_tflops": 0.7425580954,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 51,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 832.7109375,
"mean_tflops": 0.746142255144,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 52,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 836.7109375,
"mean_tflops": 0.749726414888,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 53,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 840.7109375,
"mean_tflops": 0.753310574632,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 54,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 844.7109375,
"mean_tflops": 0.756894734376,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 55,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 848.7109375,
"mean_tflops": 0.76047889412,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 56,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 852.7109375,
"mean_tflops": 0.764063053864,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 57,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 856.7109375,
"mean_tflops": 0.767647213608,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 58,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 860.7109375,
"mean_tflops": 0.771231373352,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 59,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 864.7109375,
"mean_tflops": 0.774815533096,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 60,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 868.7109375,
"mean_tflops": 0.77839969284,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 61,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 872.7109375,
"mean_tflops": 0.781983852584,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 62,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 876.7109375,
"mean_tflops": 0.785568012328,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 63,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 880.7109375,
"mean_tflops": 0.789152172072,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 64,
"solved": 2040,
"num_puzzles": 2048,
"solve_rate": 0.99609375,
"mean_nfe": 884.7109375,
"mean_tflops": 0.792736331816,
"accepted": 2046,
"unresolved": 2
},
{
"breadth_cap": 65,
"solved": 2041,
"num_puzzles": 2048,
"solve_rate": 0.99658203125,
"mean_nfe": 886.9140625,
"mean_tflops": 0.7947104198,
"accepted": 2047,
"unresolved": 1
},
{
"breadth_cap": 66,
"solved": 2041,
"num_puzzles": 2048,
"solve_rate": 0.99658203125,
"mean_nfe": 888.9140625,
"mean_tflops": 0.796502499672,
"accepted": 2047,
"unresolved": 1
},
{
"breadth_cap": 67,
"solved": 2041,
"num_puzzles": 2048,
"solve_rate": 0.99658203125,
"mean_nfe": 890.9140625,
"mean_tflops": 0.798294579544,
"accepted": 2047,
"unresolved": 1
},
{
"breadth_cap": 68,
"solved": 2041,
"num_puzzles": 2048,
"solve_rate": 0.99658203125,
"mean_nfe": 892.9140625,
"mean_tflops": 0.800086659416,
"accepted": 2047,
"unresolved": 1
},
{
"breadth_cap": 69,
"solved": 2041,
"num_puzzles": 2048,
"solve_rate": 0.99658203125,
"mean_nfe": 894.9140625,
"mean_tflops": 0.801878739288,
"accepted": 2047,
"unresolved": 1
},
{
"breadth_cap": 70,
"solved": 2041,
"num_puzzles": 2048,
"solve_rate": 0.99658203125,
"mean_nfe": 896.9140625,
"mean_tflops": 0.80367081916,
"accepted": 2047,
"unresolved": 1
},
{
"breadth_cap": 71,
"solved": 2041,
"num_puzzles": 2048,
"solve_rate": 0.99658203125,
"mean_nfe": 898.9140625,
"mean_tflops": 0.805462899032,
"accepted": 2047,
"unresolved": 1
},
{
"breadth_cap": 72,
"solved": 2042,
"num_puzzles": 2048,
"solve_rate": 0.9970703125,
"mean_nfe": 899.75,
"mean_tflops": 0.806211932416,
"accepted": 2048,
"unresolved": 0
}
],
"selection_rule": "stop each trajectory once the trailing-mean residual remains <= threshold for 8 checks and the decoded grid remains unchanged; restart only trajectories that reach max_nfe without acceptance; if max_breadth is exhausted, return the attempted terminal prediction with minimum residual",
"labels_used_for": "final scoring only; never stopping, restart, or candidate selection"
}