{ "bench_version": 1, "track": "dynamics", "model": "custom_switched_observer_rnn", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.006, "residual_scale": 0.0 }, "sweep": [ { "cfg": { "lr": 0.0015, "residual_scale": 0.0 }, "mean": 0.013166054850444198 }, { "cfg": { "lr": 0.0015, "residual_scale": 0.45 }, "mean": 0.014295766479335725 }, { "cfg": { "lr": 0.0015, "residual_scale": 0.9 }, "mean": 0.0171925718896091 }, { "cfg": { "lr": 0.003, "residual_scale": 0.0 }, "mean": 0.004933564574457705 }, { "cfg": { "lr": 0.003, "residual_scale": 0.45 }, "mean": 0.0055945796193555 }, { "cfg": { "lr": 0.003, "residual_scale": 0.9 }, "mean": 0.00661358906654641 }, { "cfg": { "lr": 0.006, "residual_scale": 0.0 }, "mean": 0.002542233618441969 }, { "cfg": { "lr": 0.006, "residual_scale": 0.45 }, "mean": 0.002950671303551644 }, { "cfg": { "lr": 0.006, "residual_scale": 0.9 }, "mean": 0.0036307289847172797 } ], "full": { "mean": 0.0026320953911636025, "std": 0.0005285747590703118, "per_seed": [ 0.0022362906020134687, 0.0019099744968116283, 0.002377047436311841, 0.0036456219386309385, 0.0032889561261981726, 0.002569406060501933, 0.0024848924949765205, 0.002544573973864317 ], "n": 8 } }, "idea": { "mean": 0.0031172324815997854, "std": 0.0013738510956345731, "per_seed": [ 0.003699983935803175, 0.002306041307747364, 0.0014978268882259727, 0.004068186040967703, 0.0020008618012070656, 0.0021134966518729925, 0.0032527041621506214, 0.005998759064823389 ], "n": 8 }, "comparison": { "delta_mean": 0.0004851370904361829, "idea_wins": 3, "n_pairs": 8, "per_seed_diffs": [ 0.0014636933337897062, 0.0003960668109357357, -0.0008792205480858684, 0.00042256410233676434, -0.001288094324991107, -0.0004559094086289406, 0.0007678116671741009, 0.003454185090959072 ], "p_value": 0.43845, "mde": 0.0012544680223890213, "mde_rel_pct": 47.6604315558048, "verdict": "no significant win", "system_worked": false }, "mechanism_signature": { "prediction": "cycle sensitivity should be below individual branch sensitivity", "observed_mean_branch_norms": [ 1.3430568054318428, 1.3305060217777889 ], "observed_cycle_norm_product": 1.786945167216707, "confirmed": false }, "idea_sweep": [ { "cfg": { "lr": 0.006, "gain": 0.45 }, "mean_on_sweep_seeds": 0.0028930095431860536 }, { "cfg": { "lr": 0.0015, "gain": 0.45 }, "mean_on_sweep_seeds": 0.01692552154418081 }, { "cfg": { "lr": 0.006, "gain": 0.45 }, "mean_on_sweep_seeds": 0.0028930095431860536 } ] }