{
  "bench_version": 1,
  "track": "dynamics",
  "model": "rnn_small",
  "metric_direction": "lower is better",
  "n_seeds": 8,
  "baseline": {
    "best_cfg": {
      "lr": 0.01
    },
    "sweep": [
      {
        "cfg": {
          "lr": 0.001
        },
        "mean": 0.4970664903521538
      },
      {
        "cfg": {
          "lr": 0.003
        },
        "mean": 0.06318101449869573
      },
      {
        "cfg": {
          "lr": 0.01
        },
        "mean": 0.011876139731612056
      }
    ],
    "full": {
      "mean": 0.011876139731612056,
      "std": 0.00487020680273246,
      "per_seed": [
        0.010697474703192711,
        0.011668417602777481,
        0.012986297719180584,
        0.019935088232159615,
        0.005051083862781525,
        0.009833427146077156,
        0.018368331715464592,
        0.006468996871262789
      ],
      "n": 8
    }
  },
  "idea": {
    "mean": 0.0054059389512985945,
    "std": 0.001747436411673786,
    "per_seed": [
      0.007179650943726301,
      0.003484317334368825,
      0.006277043838053942,
      0.004052966833114624,
      0.004533923696726561,
      0.0029652707744389772,
      0.007882867939770222,
      0.006871470250189304
    ],
    "n": 8,
    "best_cfg": {
      "lr": 0.01
    }
  },
  "comparison": {
    "delta_mean": -0.006470200780313462,
    "idea_wins": 7,
    "n_pairs": 8,
    "per_seed_diffs": [
      -0.0035178237594664097,
      -0.008184100268408656,
      -0.006709253881126642,
      -0.01588212139904499,
      -0.0005171601660549641,
      -0.006868156371638179,
      -0.01048546377569437,
      0.0004024733789265156
    ],
    "p_value": 0.0156,
    "mde": 0.004461024219396654,
    "mde_rel_pct": 37.56291455145348,
    "verdict": "idea better (significant)",
    "system_worked": true
  },
  "baseline_grid": [
    {
      "lr": 0.001
    },
    {
      "lr": 0.003
    },
    {
      "lr": 0.01
    }
  ],
  "idea_grid": [
    {
      "cfg": {
        "lr": 0.001
      },
      "result": {
        "mean": 0.1896328409202397,
        "std": 0.10288670070387347,
        "per_seed": [
          0.3404313623905182,
          0.22487185895442963,
          0.07244019210338593,
          0.19320537149906158,
          0.2565637230873108,
          0.05929507687687874,
          0.07232630252838135,
          0.2979288399219513
        ],
        "n": 8
      }
    },
    {
      "cfg": {
        "lr": 0.003
      },
      "result": {
        "mean": 0.02868673810735345,
        "std": 0.012694117890344024,
        "per_seed": [
          0.04536343365907669,
          0.028771521523594856,
          0.02157784253358841,
          0.036481499671936035,
          0.03701217100024223,
          0.00805759523063898,
          0.012121920473873615,
          0.04010792076587677
        ],
        "n": 8
      }
    },
    {
      "cfg": {
        "lr": 0.01
      },
      "result": {
        "mean": 0.0054059389512985945,
        "std": 0.001747436411673786,
        "per_seed": [
          0.007179650943726301,
          0.003484317334368825,
          0.006277043838053942,
          0.004052966833114624,
          0.004533923696726561,
          0.0029652707744389772,
          0.007882867939770222,
          0.006871470250189304
        ],
        "n": 8,
        "best_cfg": {
          "lr": 0.01
        }
      }
    }
  ],
  "math_check_exponential_residual": 1.9421449906076305e-06,
  "protocol_note": "dynamics selected because the idea claims stability/control through norm-preserving recurrent transformations",
  "mechanism_signature": {
    "prediction": "KLU recurrent map preserves the norm of each incoming hidden state while dense recurrence need not",
    "observed_idea_recurrent_norm": 2.192119225859642,
    "observed_baseline_recurrent_output_norm": 1.5108804553747177,
    "observed_idea_unitarity_residual": 1.8216120736269659e-06,
    "confirmed": true
  }
}