{
  "bench_version": 1,
  "track": "dynamics",
  "model": "rnn_small",
  "metric_direction": "lower is better",
  "n_seeds": 8,
  "baseline": {
    "best_cfg": {
      "lr": 0.01,
      "epochs": 12
    },
    "sweep": [
      {
        "cfg": {
          "lr": 0.001,
          "epochs": 12
        },
        "per_seed": [
          0.0070558576844632626,
          0.0025721259880810976,
          0.0027294710744172335,
          0.0034412096720188856,
          0.0046035656705498695,
          0.003881418611854315,
          0.003549819579347968,
          0.008237835951149464
        ],
        "mean": 0.004508913028985262,
        "std": 0.002062311810243919
      },
      {
        "cfg": {
          "lr": 0.003,
          "epochs": 12
        },
        "per_seed": [
          0.0019035028526559472,
          0.0020239646546542645,
          0.0036072093062102795,
          0.006420358084142208,
          0.0027512782253324986,
          0.0025861035101115704,
          0.0030072720255702734,
          0.0019430352840572596
        ],
        "mean": 0.0030303404928417876,
        "std": 0.0014911902872439856
      },
      {
        "cfg": {
          "lr": 0.01,
          "epochs": 12
        },
        "per_seed": [
          0.0018195160664618015,
          0.0012350792530924082,
          0.0008721873164176941,
          0.0027302715461701155,
          0.0014946963638067245,
          0.0011278137098997831,
          0.0009376407833769917,
          0.002133276779204607
        ],
        "mean": 0.0015438102273037657,
        "std": 0.0006459183721555687
      }
    ],
    "harness_tuning": {
      "best_cfg": {
        "lr": 0.01,
        "epochs": 12
      },
      "sweep": [
        {
          "cfg": {
            "lr": 0.001,
            "epochs": 12
          },
          "mean": 0.00394966610474512
        },
        {
          "cfg": {
            "lr": 0.003,
            "epochs": 12
          },
          "mean": 0.003488758724415675
        },
        {
          "cfg": {
            "lr": 0.01,
            "epochs": 12
          },
          "mean": 0.0016642635455355048
        }
      ],
      "full": {
        "mean": 0.0015438102273037657,
        "std": 0.0006042013121072209,
        "per_seed": [
          0.0018195160664618015,
          0.0012350792530924082,
          0.0008721873164176941,
          0.0027302715461701155,
          0.0014946963638067245,
          0.0011278137098997831,
          0.0009376407833769917,
          0.002133276779204607
        ],
        "n": 8
      }
    },
    "full": {
      "per_seed": [
        0.0018195160664618015,
        0.0012350792530924082,
        0.0008721873164176941,
        0.0027302715461701155,
        0.0014946963638067245,
        0.0011278137098997831,
        0.0009376407833769917,
        0.002133276779204607
      ],
      "mean": 0.0015438102273037657,
      "std": 0.0006459183721555687
    }
  },
  "idea": {
    "per_seed": [
      0.010384509339928627,
      0.014125057496130466,
      0.052780672907829285,
      0.03464017063379288,
      0.008677596226334572,
      0.00931015145033598,
      0.0071034664288163185,
      0.005731561686843634
    ],
    "mean": 0.01784414827125147,
    "std": 0.016865335320885074
  },
  "comparison": {
    "delta_mean": 0.016300338043947704,
    "idea_wins": 0,
    "n_pairs": 8,
    "per_seed_diffs": [
      0.008564993273466825,
      0.012889978243038058,
      0.05190848559141159,
      0.03190989908762276,
      0.007182899862527847,
      0.008182337740436196,
      0.006165825645439327,
      0.0035982849076390266
    ],
    "p_value": 0.0081,
    "mde": 0.014119949734496392,
    "mde_rel_pct": 914.6169318463842,
    "verdict": "idea worse (significant)",
    "system_worked": false
  },
  "mechanism_signature": {
    "idea_sweep": [
      {
        "cfg": {
          "lr": 0.001,
          "epochs": 12
        },
        "per_seed": [
          0.004393833689391613,
          0.02482367865741253,
          0.0714903250336647,
          0.013632673770189285,
          0.0009164400398731232,
          0.009770718403160572,
          0.024457935243844986,
          0.019280623644590378
        ],
        "mean": 0.0210957785602659,
        "std": 0.02216245868447279
      },
      {
        "cfg": {
          "lr": 0.003,
          "epochs": 12
        },
        "per_seed": [
          0.01244353037327528,
          0.03397444263100624,
          0.05305279791355133,
          0.0075877052731812,
          0.0036976668052375317,
          0.003634050488471985,
          0.02957206219434738,
          0.011668341234326363
        ],
        "mean": 0.019453824614174664,
        "std": 0.01769262402163577
      },
      {
        "cfg": {
          "lr": 0.01,
          "epochs": 12
        },
        "per_seed": [
          0.010384509339928627,
          0.014125057496130466,
          0.052780672907829285,
          0.03464017063379288,
          0.008677596226334572,
          0.00931015145033598,
          0.0071034664288163185,
          0.005731561686843634
        ],
        "mean": 0.01784414827125147,
        "std": 0.016865335320885074
      }
    ],
    "mechanism_signature": {
      "prediction": "Cayley preserves lifted norm; explicit Euler grows it",
      "observed_cayley_relative_norm_error": 5.960464477539062e-07,
      "observed_euler_20_step_norm_ratio": 1.003595232963562,
      "trained_test_mse": 0.01244353037327528,
      "confirmed": true
    }
  },
  "custom_track": null
}