{ "bench_version": 1, "track": "sequence", "model": "transformer_tiny", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.001 }, "sweep": [ { "cfg": { "lr": 0.001 }, "mean": 0.47025539726018906 }, { "cfg": { "lr": 0.003 }, "mean": 0.5620733425021172 }, { "cfg": { "lr": 0.006 }, "mean": 0.7826127856969833 } ], "full": { "mean": 0.482548501342535, "std": 0.03354559876849172, "per_seed": [ 0.4842072129249573, 0.42341384291648865, 0.4566255807876587, 0.5167749524116516, 0.49487417936325073, 0.4824545681476593, 0.5386028289794922, 0.4634348452091217 ], "n": 8 } }, "idea": { "mean": 0.4949374906718731, "std": 0.053264005967373906, "per_seed": [ 0.48880892992019653, 0.4039829969406128, 0.4540945589542389, 0.5083529353141785, 0.5644499659538269, 0.5458484292030334, 0.5475071668624878, 0.4464549422264099 ], "n": 8 }, "comparison": { "delta_mean": 0.012388989329338074, "idea_wins": 4, "n_pairs": 8, "per_seed_diffs": [ 0.004601716995239258, -0.019430845975875854, -0.0025310218334198, -0.008422017097473145, 0.06957578659057617, 0.06339386105537415, 0.008904337882995605, -0.016979902982711792 ], "p_value": 0.42645, "mde": 0.02909483793289158, "mde_rel_pct": 6.029412142394932, "verdict": "no significant win", "system_worked": false }, "track_match": "sequence-level correlated multi-token forecast", "idea_best_cfg": { "lr": 0.001 }, "idea_grid": [ { "cfg": { "lr": 0.001 }, "full": { "mean": 0.4949374906718731, "std": 0.053264005967373906, "per_seed": [ 0.48880892992019653, 0.4039829969406128, 0.4540945589542389, 0.5083529353141785, 0.5644499659538269, 0.5458484292030334, 0.5475071668624878, 0.4464549422264099 ], "n": 8 } }, { "cfg": { "lr": 0.003 }, "full": { "mean": 0.6153423972427845, "std": 0.08454027758325999, "per_seed": [ 0.5847313404083252, 0.42449966073036194, 0.6188046336174011, 0.6627813577651978, 0.7065386772155762, 0.6074619889259338, 0.7107876539230347, 0.6071338653564453 ], "n": 8 } }, { "cfg": { "lr": 0.006 }, "full": { "mean": 0.8792074918746948, "std": 0.3064714056914715, "per_seed": [ 0.6058098673820496, 0.64018714427948, 0.6481318473815918, 0.9747511148452759, 1.0847872495651245, 0.8413496017456055, 1.5654103755950928, 0.6732327342033386 ], "n": 8 } } ], "epochs": 10, "n_train": 400, "n_test": 200, "mechanism_signature": { "mean_baseline_output_noise_ratio": 0.17996257916092873, "mean_idea_output_noise_ratio": 0.16205820441246033, "predicted": "idea should attenuate feature perturbation", "confirmed": true, "per_seed": [ { "baseline_output_noise_ratio": 0.09764544665813446, "idea_output_noise_ratio": 0.1503368318080902, "baseline_metric": 0.4308149516582489, "idea_metric": 0.5847313404083252 }, { "baseline_output_noise_ratio": 0.14380529522895813, "idea_output_noise_ratio": 0.1423303335905075, "baseline_metric": 0.4464939832687378, "idea_metric": 0.42449966073036194 }, { "baseline_output_noise_ratio": 0.2634066641330719, "idea_output_noise_ratio": 0.2061924785375595, "baseline_metric": 0.6809554696083069, "idea_metric": 0.6188046336174011 }, { "baseline_output_noise_ratio": 0.20416639745235443, "idea_output_noise_ratio": 0.15186598896980286, "baseline_metric": 0.690028965473175, "idea_metric": 0.6627813577651978 }, { "baseline_output_noise_ratio": 0.13947272300720215, "idea_output_noise_ratio": 0.1979530155658722, "baseline_metric": 0.7755821347236633, "idea_metric": 0.7065386772155762 }, { "baseline_output_noise_ratio": 0.21130000054836273, "idea_output_noise_ratio": 0.1443224996328354, "baseline_metric": 0.7781144380569458, "idea_metric": 0.6074619889259338 }, { "baseline_output_noise_ratio": 0.2646835446357727, "idea_output_noise_ratio": 0.11280205845832825, "baseline_metric": 0.655863881111145, "idea_metric": 0.7107876539230347 }, { "baseline_output_noise_ratio": 0.1152205616235733, "idea_output_noise_ratio": 0.1906624287366867, "baseline_metric": 0.6570667028427124, "idea_metric": 0.6071338653564453 } ] } }