{ "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.006, "weight_decay": 0.0001 }, "sweep": [ { "cfg": { "lr": 0.001, "weight_decay": 0.0 }, "mean": 0.004298779065720737 }, { "cfg": { "lr": 0.001, "weight_decay": 0.0001 }, "mean": 0.0042995543335564435 }, { "cfg": { "lr": 0.003, "weight_decay": 0.0 }, "mean": 0.0034615940821822733 }, { "cfg": { "lr": 0.003, "weight_decay": 0.0001 }, "mean": 0.003432174737099558 }, { "cfg": { "lr": 0.006, "weight_decay": 0.0 }, "mean": 0.0018282315868418664 }, { "cfg": { "lr": 0.006, "weight_decay": 0.0001 }, "mean": 0.0017760894843377173 } ], "full": { "mean": 0.0020704464986920357, "std": 0.00039353273372166414, "per_seed": [ 0.0019586661364883184, 0.0013749259524047375, 0.00156879099085927, 0.002201974857598543, 0.0023778362665325403, 0.002474326640367508, 0.0020717033185064793, 0.0025353478267788887 ], "n": 8 } }, "idea": { "mean": 0.0014319315523607656, "std": 0.00040801966999132736, "per_seed": [ 0.00139222526922822, 0.0008285212097689509, 0.0013016352895647287, 0.001789333182387054, 0.000968311564065516, 0.0018908914644271135, 0.0012502360623329878, 0.002034298377111554 ], "n": 8 }, "comparison": { "delta_mean": -0.00063851494633127, "idea_wins": 8, "n_pairs": 8, "per_seed_diffs": [ -0.0005664408672600985, -0.0005464047426357865, -0.00026715570129454136, -0.0004126416752114892, -0.0014095247024670243, -0.0005834351759403944, -0.0008214672561734915, -0.0005010494496673346 ], "p_value": 0.0081, "mde": 0.0002917931104320964, "mde_rel_pct": 14.093245617137706, "verdict": "idea better (significant)", "system_worked": true }, "mechanism_signature": { "predicted_vs_observed": { "target_radius": 0.9, "observed_radius_mean": 1.5063634291291237, "fitted_mode_rows": [ { "epoch": 0, "fitted_radius": 0.8541139240506328, "fitted_omega": 0.05 }, { "epoch": 11, "fitted_radius": 0.894367088607595, "fitted_omega": 0.05 } ] }, "confirmed": false }, "idea_trials": [ { "cfg": { "lr": 0.006, "weight_decay": 0.0001, "r_max": 0.9, "beta": 0.02 }, "result": { "mean": 0.0014319315523607656, "std": 0.00040801966999132736, "per_seed": [ 0.00139222526922822, 0.0008285212097689509, 0.0013016352895647287, 0.001789333182387054, 0.000968311564065516, 0.0018908914644271135, 0.0012502360623329878, 0.002034298377111554 ], "n": 8 } }, { "cfg": { "lr": 0.003, "weight_decay": 0.0001, "r_max": 0.9, "beta": 0.02 }, "result": { "mean": 0.0028593493334483355, "std": 0.001481906654941589, "per_seed": [ 0.0025175209157168865, 0.0032294574193656445, 0.0012714689364656806, 0.006308244541287422, 0.002197133842855692, 0.003583323210477829, 0.0020109813194721937, 0.0017566644819453359 ], "n": 8 } }, { "cfg": { "lr": 0.006, "weight_decay": 0.0001, "r_max": 0.9, "beta": 0.02 }, "result": { "mean": 0.0014319315523607656, "std": 0.00040801966999132736, "per_seed": [ 0.00139222526922822, 0.0008285212097689509, 0.0013016352895647287, 0.001789333182387054, 0.000968311564065516, 0.0018908914644271135, 0.0012502360623329878, 0.002034298377111554 ], "n": 8 } } ], "math_sanity": { "true_radius": 0.93, "fitted_radius": 0.9346202531645569, "true_omega": 0.42, "fitted_omega": 0.4275280898876404, "radius_abs_error": 0.004620253164556831, "frequency_abs_error": 0.007528089887640421, "pass": true } }