{ "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.01 }, "sweep": [ { "cfg": { "lr": 0.001 }, "mean": 0.005056460213381797 }, { "cfg": { "lr": 0.003 }, "mean": 0.0029833840089850128 }, { "cfg": { "lr": 0.01 }, "mean": 0.0014822854354861192 } ], "full": { "mean": 0.0014822854354861192, "std": 0.0010530387951197904, "per_seed": [ 0.00411539152264595, 0.0017662415048107505, 0.0006022244924679399, 0.0013981220545247197, 0.0012625530362129211, 0.0010164444101974368, 0.0009094251436181366, 0.000787881319411099 ], "n": 8 } }, "idea": { "mean": 0.0021019700725446455, "std": 0.0007477001545769394, "per_seed": [ 0.0022576870396733284, 0.0022740764543414116, 0.0024945440236479044, 0.0014605424366891384, 0.0016900348709896207, 0.0007099835784174502, 0.0026052906177937984, 0.003323601558804512 ], "n": 8, "cfg": { "lr": 0.01, "rank": 40 }, "sweep": [ { "cfg": { "lr": 0.01, "rank": 24 }, "mean": 0.003177800477715209, "std": 0.0010355345041122954, "per_seed": [ 0.0036525779869407415, 0.004798321519047022, 0.004517960362136364, 0.002177618443965912, 0.0021414996590465307, 0.0027073805686086416, 0.003489549970254302, 0.0019374953117221594 ], "n": 8 }, { "cfg": { "lr": 0.01, "rank": 32 }, "mean": 0.002437932969769463, "std": 0.0011340391219749532, "per_seed": [ 0.0013505988754332066, 0.0022953480947762728, 0.002628838177770376, 0.002794082509353757, 0.0051675527356565, 0.001575573580339551, 0.0017304965294897556, 0.0019609732553362846 ], "n": 8 }, { "cfg": { "lr": 0.01, "rank": 40 }, "mean": 0.0021019700725446455, "std": 0.0007477001545769394, "per_seed": [ 0.0022576870396733284, 0.0022740764543414116, 0.0024945440236479044, 0.0014605424366891384, 0.0016900348709896207, 0.0007099835784174502, 0.0026052906177937984, 0.003323601558804512 ], "n": 8 } ] }, "comparison": { "delta_mean": 0.0006196846370585263, "idea_wins": 2, "n_pairs": 8, "per_seed_diffs": [ -0.001857704482972622, 0.0005078349495306611, 0.0018923195311799645, 6.24203821644187e-05, 0.00042748183477669954, -0.0003064608317799866, 0.0016958654741756618, 0.002535720239393413 ], "p_value": 0.20345, "mde": 0.0011741999248702343, "mde_rel_pct": 79.21550713241356, "verdict": "no significant win", "system_worked": false, "paired_delta": [ -0.001857704482972622, 0.0005078349495306611, 0.0018923195311799645, 6.24203821644187e-05, 0.00042748183477669954, -0.0003064608317799866, 0.0016958654741756618, 0.002535720239393413 ] }, "mechanism_signature": { "predicted_state_cost_ratio": 0.625, "observed_state_cost_ratio": 0.625, "baseline_hidden": 64, "idea_hidden": 40, "baseline_mean_pred_observed_corr": 0.9994533416434941, "idea_mean_pred_observed_corr": 0.9989892089422736, "confirmed": true } }