{
  "bench_version": 1,
  "track": "tabular",
  "model": "mlp_tiny",
  "metric_direction": "lower is better",
  "n_seeds": 8,
  "baseline": {
    "best_cfg": {
      "lr": 0.003,
      "temperature": 0.0005
    },
    "sweep": [
      {
        "cfg": {
          "lr": 0.001,
          "temperature": 0.0005
        },
        "mean": 6.877625823020935
      },
      {
        "cfg": {
          "lr": 0.001,
          "temperature": 0.002
        },
        "mean": 6.878889203071594
      },
      {
        "cfg": {
          "lr": 0.003,
          "temperature": 0.0005
        },
        "mean": 6.73323667049408
      },
      {
        "cfg": {
          "lr": 0.003,
          "temperature": 0.002
        },
        "mean": 6.734855055809021
      },
      {
        "cfg": {
          "lr": 0.006,
          "temperature": 0.0005
        },
        "mean": 7.187614321708679
      },
      {
        "cfg": {
          "lr": 0.006,
          "temperature": 0.002
        },
        "mean": 7.431662321090698
      }
    ],
    "full": {
      "mean": 6.753931879997253,
      "std": 0.15501289580001476,
      "per_seed": [
        6.430861949920654,
        6.871230602264404,
        6.933614253997803,
        6.697239875793457,
        6.64493465423584,
        6.860416412353516,
        6.7199554443359375,
        6.873201847076416
      ],
      "n": 8
    }
  },
  "idea": {
    "mean": 6.753931879997253,
    "std": 0.15501289580001476,
    "per_seed": [
      6.430861949920654,
      6.871230602264404,
      6.933614253997803,
      6.697239875793457,
      6.64493465423584,
      6.860416412353516,
      6.7199554443359375,
      6.873201847076416
    ],
    "n": 8
  },
  "comparison": {
    "delta_mean": 0.0,
    "idea_wins": 0,
    "n_pairs": 8,
    "per_seed_diffs": [
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0
    ],
    "p_value": 1.0,
    "mde": 0.0,
    "mde_rel_pct": 0.0,
    "verdict": "no measurable effect",
    "system_worked": false
  },
  "idea_sweep": [
    [
      6.877625823020935,
      {
        "lr": 0.001,
        "temperature": 0.0005
      }
    ],
    [
      6.878889203071594,
      {
        "lr": 0.001,
        "temperature": 0.002
      }
    ],
    [
      6.73323667049408,
      {
        "lr": 0.003,
        "temperature": 0.0005
      }
    ]
  ],
  "protocol_note": "Official bench tabular track; custom Langevin loop is the intervention.",
  "mechanism_signature": {
    "prediction": "observed unsafe frequency should not exceed certificate envelope",
    "predicted_final_bound": 1.0,
    "observed_mean_unsafe_frequency": 1.0,
    "observed_post_stop_unsafe_frequency": 1.0,
    "n_trained_models": 8,
    "confirmed": true
  }
}