{ "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.006, "epochs": 12 }, "sweep": [ { "cfg": { "lr": 0.001, "epochs": 12 }, "mean": 0.010470356093719602 }, { "cfg": { "lr": 0.003, "epochs": 12 }, "mean": 0.005202302592806518 }, { "cfg": { "lr": 0.006, "epochs": 12 }, "mean": 0.0028439466550480574 } ], "full": { "per_seed": [ 0.0017406052211299539, 0.0037223200779408216, 0.002092725830152631, 0.0038201354909688234, 0.0038962536491453648, 0.0017676886636763811, 0.0043351962231099606, 0.0036920576822012663 ], "mean": 0.0031333728547906503, "signatures": [ { "edge_mean": 0.6060569062829018, "gain_final": 1.0 }, { "edge_mean": 0.6419745236635208, "gain_final": 1.0 }, { "edge_mean": 0.6132491379976273, "gain_final": 1.0 }, { "edge_mean": 0.6298487186431885, "gain_final": 1.0 }, { "edge_mean": 0.6433217823505402, "gain_final": 1.0 }, { "edge_mean": 0.6161705479025841, "gain_final": 1.0 }, { "edge_mean": 0.6346533372998238, "gain_final": 1.0 }, { "edge_mean": 0.6225631684064865, "gain_final": 1.0 } ] } }, "idea": { "best_cfg": { "lr": 0.006, "epochs": 12 }, "sweep": [ { "cfg": { "lr": 0.001, "epochs": 12 }, "mean": 0.013453479390591383 }, { "cfg": { "lr": 0.003, "epochs": 12 }, "mean": 0.004181064839940518 }, { "cfg": { "lr": 0.006, "epochs": 12 }, "mean": 0.0015980473981471732 } ], "per_seed": [ 0.000768338271882385, 0.0010514536406844854, 0.0033124624751508236, 0.0012599352048709989, 0.0007571351598016918, 0.0012769970344379544, 0.002315002726390958, 0.001677743741311133 ], "mean": 0.0015523835318163037, "signatures": [ { "edge_mean": 0.8978222989738791, "gain_final": 1.507260824388018 }, { "edge_mean": 0.8991352558280741, "gain_final": 1.4971718810377141 }, { "edge_mean": 0.9008403422532384, "gain_final": 1.4441157022221471 }, { "edge_mean": 0.9019607306183688, "gain_final": 1.4800354971216234 }, { "edge_mean": 0.9020658948006778, "gain_final": 1.4013476220899848 }, { "edge_mean": 0.9014030846181611, "gain_final": 1.4688920100836202 }, { "edge_mean": 0.8958116604630089, "gain_final": 1.365893652127335 }, { "edge_mean": 0.8992522999588555, "gain_final": 1.5118891100008007 } ] }, "comparison": { "delta_mean": -0.0015809893229743466, "per_seed_diffs": [ -0.0009722669492475688, -0.002670866437256336, 0.0012197366449981928, -0.0025602002860978246, -0.003139118489343673, -0.0004906916292384267, -0.0020201934967190027, -0.0020143139408901334 ], "idea_wins": 7, "n_pairs": 8, "p_value": 0.0324, "verdict": "idea better (significant)" }, "mechanism_signature": { "prediction": "controlled effective edge near target below 1", "predicted_target": 0.9, "observed_edge_mean": 0.899786445939283, "relative_error": 0.0002372822896855463, "confirmed": true } }