{ "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.005 }, "sweep": [ { "cfg": { "lr": 0.001 }, "mean": 0.0014035784406587481 }, { "cfg": { "lr": 0.003 }, "mean": 0.000857145045301877 }, { "cfg": { "lr": 0.005 }, "mean": 0.000637838282273151 } ], "full": { "mean": 0.0006355784062179737, "std": 0.00018562462201799073, "per_seed": [ 0.0005684797652065754, 0.0006359252729453146, 0.0004176601651124656, 0.0009292879258282483, 0.0004875308950431645, 0.0004790188977494836, 0.0006295977509580553, 0.0009371265769004822 ], "n": 8 }, "parity_full": [ { "cfg": { "lr": 0.001 }, "full": { "mean": 0.0016591226740274578, "std": 0.0005884083669493902, "per_seed": [ 0.0011428100988268852, 0.001511492533609271, 0.001185992150567472, 0.0017740189796313643, 0.0012394130462780595, 0.0013388715451583266, 0.002085721120238304, 0.00299466191790998 ], "n": 8 } }, { "cfg": { "lr": 0.003 }, "full": { "mean": 0.0009047073835972697, "std": 0.00019738432718009543, "per_seed": [ 0.0007076333276927471, 0.0008893113117665052, 0.0007631920161657035, 0.001068443525582552, 0.0006572386482730508, 0.0008406317210756242, 0.0010242111748084426, 0.0012869973434135318 ], "n": 8 } }, { "cfg": { "lr": 0.005 }, "full": { "mean": 0.0006355784062179737, "std": 0.00018562462201799073, "per_seed": [ 0.0005684797652065754, 0.0006359252729453146, 0.0004176601651124656, 0.0009292879258282483, 0.0004875308950431645, 0.0004790188977494836, 0.0006295977509580553, 0.0009371265769004822 ], "n": 8 } } ] }, "idea": { "mean": 0.05243536550551653, "std": 0.008883196341761043, "per_seed": [ 0.04108409211039543, 0.05810997262597084, 0.05400923267006874, 0.06831105798482895, 0.05535797029733658, 0.03811589255928993, 0.053262870758771896, 0.051231835037469864 ], "n": 8 }, "comparison": { "delta_mean": 0.051799787099298555, "idea_wins": 0, "n_pairs": 8, "per_seed_diffs": [ 0.040515612345188856, 0.057474047353025526, 0.053591572504956275, 0.0673817700590007, 0.054870439402293414, 0.03763687366154045, 0.05263327300781384, 0.05029470846056938 ], "p_value": 0.0081, "mde": 0.007859816836150254, "mde_rel_pct": 1236.640005270208, "verdict": "idea worse (significant)", "system_worked": false }, "mechanism_signature": { "bound": 1.0, "n_test": 400, "interior_fraction": 0.72, "interior_identity_max_error": 0.0, "max_constraint_violation": 0.0, "mean_shaping_distance": 0.08201595395803452, "predicted_interior_error": 0.0, "predicted_max_violation": 0.0, "confirmed": true, "math_check": { "max_error_vs_slsqp": 1.8546459140785727e-08, "max_constraint_violation": 3.3306690738754696e-16, "interior_identity_max_error": 0.0, "n_reference": 90 }, "trained_baseline_mean": 0.0006355784062179737, "trained_idea_best_mean": 0.05243536550551653 }, "idea_sweep": [ { "cfg": { "lr": 0.001 }, "full": { "mean": 0.05274156527593732, "std": 0.008962000698782165, "per_seed": [ 0.04126017168164253, 0.05878905951976776, 0.054237402975559235, 0.06873083114624023, 0.05547387897968292, 0.03838273882865906, 0.053651560097932816, 0.05140687897801399 ], "n": 8 } }, { "cfg": { "lr": 0.003 }, "full": { "mean": 0.05250410409644246, "std": 0.00887397537778388, "per_seed": [ 0.04117707535624504, 0.05814062058925629, 0.054112356156110764, 0.06837401539087296, 0.055376794189214706, 0.03818747028708458, 0.05337858572602272, 0.051285915076732635 ], "n": 8 } }, { "cfg": { "lr": 0.005 }, "full": { "mean": 0.05243536550551653, "std": 0.008883196341761043, "per_seed": [ 0.04108409211039543, 0.05810997262597084, 0.05400923267006874, 0.06831105798482895, 0.05535797029733658, 0.03811589255928993, 0.053262870758771896, 0.051231835037469864 ], "n": 8 } } ], "runtime_sec": 95.81617069244385 }