{ "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 7, "baseline": { "best_cfg": { "lr": 0.006, "knob": 0.01 }, "sweep": [ { "cfg": { "lr": 0.001, "knob": 0.01 }, "mean": 0.01457586046308279 }, { "cfg": { "lr": 0.001, "knob": 0.03 }, "mean": 0.01469542644917965 }, { "cfg": { "lr": 0.001, "knob": 0.06 }, "mean": 0.0148483463563025 }, { "cfg": { "lr": 0.003, "knob": 0.01 }, "mean": 0.0047323108301498 }, { "cfg": { "lr": 0.003, "knob": 0.03 }, "mean": 0.00487289740704 }, { "cfg": { "lr": 0.003, "knob": 0.06 }, "mean": 0.00508440809790045 }, { "cfg": { "lr": 0.006, "knob": 0.01 }, "mean": 0.0026383594959042966 }, { "cfg": { "lr": 0.006, "knob": 0.03 }, "mean": 0.002871918957680464 }, { "cfg": { "lr": 0.006, "knob": 0.06 }, "mean": 0.0032478051725775003 } ], "full": { "mean": 0.002745348698226735, "std": 0.0005053976729980165, "per_seed": [ 0.0029019152279943228, 0.002326412359252572, 0.00219854642637074, 0.0031265639699995518, 0.0031316187232732773, 0.0024797292426228523, 0.0021466391626745462, 0.0036513644736260176 ], "n": 8 } }, "idea": { "mean": 0.0027424886024424006, "std": 0.0004920238808410106, "per_seed": [ 0.002906925044953823, 0.002241562819108367, 0.0021613340359181166, 0.003099968424066901, 0.0030245676171034575, 0.002236209576949477, 0.003526852698996663 ], "n": 7 }, "comparison": { "delta_mean": 0.00012657072927270616, "idea_wins": 5, "n_pairs": 7, "per_seed_diffs": [ 5.009816959500313e-06, -8.48495401442051e-05, -3.721239045262337e-05, -2.6595545932650566e-05, -0.00010705110616981983, -0.00024351966567337513, 0.0013802135363221169 ], "p_value": 0.98495, "mde": 0.0004994211111627109, "mde_rel_pct": 18.191536524496687, "verdict": "no significant win", "system_worked": false }, "mechanism_signature": { "idea_sweep": [ { "cfg": { "lr": 0.001, "knob": 0.35 }, "mean": 0.014552626991644502 }, { "cfg": { "lr": 0.001, "knob": 0.7 }, "mean": 0.014597104163840413 }, { "cfg": { "lr": 0.001, "knob": 1.4 }, "mean": 0.014682337641716003 }, { "cfg": { "lr": 0.003, "knob": 0.35 }, "mean": 0.004710524983238429 }, { "cfg": { "lr": 0.003, "knob": 0.7 }, "mean": 0.004761852149385959 }, { "cfg": { "lr": 0.003, "knob": 1.4 }, "mean": 0.004872480174526572 }, { "cfg": { "lr": 0.006, "knob": 0.35 }, "mean": 0.002602447581011802 }, { "cfg": { "lr": 0.006, "knob": 0.7 }, "mean": 0.002704756858292967 }, { "cfg": { "lr": 0.006, "knob": 1.4 }, "mean": 0.0029094331548549235 } ], "mechanism_signature": { "trained_model": "ResidualRNN dynamics forecaster", "n_observations": 56, "predicted_conditional_gate_variance": "q*u*(1-u), q=2*kappa*dt*[r-r_target]+", "observed_variance_fit_slope": 1.0019951817912731, "expected_slope": 1.0, "relative_error": 0.001995181791273115, "confirmed": true } }, "custom_track": null }