{ "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": 0.0014828930143266916 }, { "cfg": { "lr": 0.003 }, "mean": 0.000868373506818898 }, { "cfg": { "lr": 0.01 }, "mean": 0.0004153431436861865 } ], "full": { "mean": 0.0003938443751394516, "std": 0.00015694612826889552, "per_seed": [ 0.0004267372714821249, 0.00025510918931104243, 0.00029441321385093033, 0.0006851129001006484, 0.00039094407111406326, 0.0005367420380935073, 0.00014661152090411633, 0.0004150847962591797 ], "n": 8 } }, "idea": { "mean": 0.01808006438659504, "std": 0.006827547286877937, "per_seed": [ 0.015054022893309593, 0.01623576134443283, 0.03221019357442856, 0.00664726784452796, 0.021471986547112465, 0.021092308685183525, 0.016446856781840324, 0.015482117421925068 ], "n": 8 }, "comparison": { "delta_mean": 0.01768622001145559, "idea_wins": 0, "n_pairs": 8, "per_seed_diffs": [ 0.014627285621827468, 0.01598065215512179, 0.03191578036057763, 0.005962154944427311, 0.0210810424759984, 0.020555566647090018, 0.016300245260936208, 0.015067032625665888 ], "p_value": 0.0081, "mde": 0.006166672867800231, "mde_rel_pct": 1565.763854216973, "verdict": "idea worse (significant)", "system_worked": false }, "mechanism_signature": { "mechanism_signature": { "prediction": "global phase is unobservable by magnitude trajectory", "predicted_max_abs_difference": 0.0, "observed_mean_abs_difference": 1.0522416005187551e-07, "operator_pair_sum_gap": 6.656185635112166e-06, "operator_sigma_min_S": 0.009892589246225832, "confirmed": true } }, "idea_sweep": [ { "cfg": { "lr": 0.01 }, "result": { "mean": 0.01808006438659504, "std": 0.006827547286877937, "per_seed": [ 0.015054022893309593, 0.01623576134443283, 0.03221019357442856, 0.00664726784452796, 0.021471986547112465, 0.021092308685183525, 0.016446856781840324, 0.015482117421925068 ], "n": 8 } }, { "cfg": { "lr": 0.001 }, "result": { "mean": 0.01906815404072404, "std": 0.004254108046330375, "per_seed": [ 0.018856093287467957, 0.01202596165239811, 0.0262176301330328, 0.024596311151981354, 0.019076775759458542, 0.01587844267487526, 0.017323212698101997, 0.018570804968476295 ], "n": 8 } }, { "cfg": { "lr": 0.003 }, "result": { "mean": 0.018212101538665593, "std": 0.0021132105733096982, "per_seed": [ 0.017741071060299873, 0.01399772334843874, 0.02053205855190754, 0.0214164387434721, 0.018905995413661003, 0.018360335379838943, 0.017668306827545166, 0.017074882984161377 ], "n": 8 } } ], "parameter_counts": { "baseline": 13313, "idea": 24961 } }