{
  "bench_version": 1,
  "track": "dynamics",
  "model": "rnn_small",
  "metric_direction": "lower is better",
  "n_seeds": 8,
  "baseline": {
    "best_cfg": {
      "lr": 0.01,
      "weight_decay": 0.0
    },
    "sweep": [
      {
        "cfg": {
          "lr": 0.001,
          "weight_decay": 0.0
        },
        "mean": 0.004298779065720737
      },
      {
        "cfg": {
          "lr": 0.003,
          "weight_decay": 0.0
        },
        "mean": 0.0034615940821822733
      },
      {
        "cfg": {
          "lr": 0.01,
          "weight_decay": 0.0
        },
        "mean": 0.0014865802077110857
      }
    ],
    "full": {
      "mean": 0.0014198302378645167,
      "std": 0.00039839114901047844,
      "per_seed": [
        0.0009327387670055032,
        0.0018641706556081772,
        0.0010785290505737066,
        0.0020708823576569557,
        0.0009260809747502208,
        0.0014055331703275442,
        0.0016443432541564107,
        0.001436363672837615
      ],
      "n": 8
    }
  },
  "idea": {
    "mean": 0.001362547576718498,
    "std": 0.00047644182050365723,
    "per_seed": [
      0.0014115098165348172,
      0.0007051810971461236,
      0.0013796384446322918,
      0.001993401674553752,
      0.001138740568421781,
      0.0022058519534766674,
      0.0010818139417096972,
      0.0009842431172728539
    ],
    "n": 8
  },
  "comparison": {
    "delta_mean": -5.7282661146018654e-05,
    "idea_wins": 4,
    "n_pairs": 8,
    "per_seed_diffs": [
      0.00047877104952931404,
      -0.0011589895584620535,
      0.00030110939405858517,
      -7.748068310320377e-05,
      0.0002126595936715603,
      0.0008003187831491232,
      -0.0005625293124467134,
      -0.00045212055556476116
    ],
    "p_value": 0.7975,
    "mde": 0.0005334826450520739,
    "mde_rel_pct": 37.573692320742076,
    "verdict": "no significant win",
    "system_worked": false
  },
  "mechanism_signature": {
    "quantity": "source-to-sink subset test MSE (trained models)",
    "baseline_predicted": 0.0010878387474804185,
    "dual_predicted": 0.0006060761043045204,
    "observed_reduction": 0.0004817626431758981,
    "prediction": "Q should reduce directed source-sink prediction error",
    "confirmed": true
  },
  "idea_sweep": [
    {
      "cfg": {
        "lr": 0.001,
        "weight_decay": 0.0
      },
      "full": {
        "mean": 0.03154140734113753,
        "std": 0.0061376140406840096,
        "per_seed": [
          0.03103465959429741,
          0.03406568616628647,
          0.028227416798472404,
          0.033552005887031555,
          0.03155306726694107,
          0.04121524840593338,
          0.01821603998541832,
          0.03446713462471962
        ],
        "n": 8
      }
    },
    {
      "cfg": {
        "lr": 0.003,
        "weight_decay": 0.0
      },
      "full": {
        "mean": 0.005806407978525385,
        "std": 0.0014531884675848988,
        "per_seed": [
          0.005496995989233255,
          0.0035544722341001034,
          0.006204786244779825,
          0.008257976733148098,
          0.0056442818604409695,
          0.003874243935570121,
          0.006419075652956963,
          0.006999431177973747
        ],
        "n": 8
      }
    },
    {
      "cfg": {
        "lr": 0.01,
        "weight_decay": 0.0
      },
      "full": {
        "mean": 0.001362547576718498,
        "std": 0.00047644182050365723,
        "per_seed": [
          0.0014115098165348172,
          0.0007051810971461236,
          0.0013796384446322918,
          0.001993401674553752,
          0.001138740568421781,
          0.0022058519534766674,
          0.0010818139417096972,
          0.0009842431172728539
        ],
        "n": 8
      }
    }
  ],
  "custom_track": null
}