# Стенд-проверка (stage-2) · промт оператора:

(универсальный)

Ответ агента:

{
  "worked": false,
  "confidence": 10,
  "verdict": "Built and ran an end-to-end paired benchmark on the structurally matched tabular Friedman regression track using the same MLP architecture, optimizer, learning rates, epochs, and seeds for ReLU and the signed log-scale activation. ReLU achieved mean test MSE 7.93194 versus 25.54624 for the idea; the paired delta was +17.61430 with permutation p=0.0081, so the idea was significantly worse. The trained-model mechanism probe was not confirmed because its increasing-magnitude probe measured the reciprocal slope ratio 0.5 rather than the expected ordered ratio 2.0.",
  "metrics": {
    "baseline": "mean test MSE 7.931941092014313, std 0.8101161551333251, best lr=0.01, epochs=20",
    "idea": "mean test MSE 25.546241760253906, std 1.397921331795095, best lr=0.01, epochs=20",
    "paired_delta": "+17.614300668239594; p=0.0081; idea wins 0/8"
  },
  "bench_report": {
    "bench_version": 1,
    "track": "tabular",
    "model": "mlp_tiny_shared_activation",
    "metric_direction": "lower is better",
    "n_seeds": 8,
    "baseline": {
      "best_cfg": {
        "lr": 0.01,
        "epochs": 20
      },
      "sweep": [
        {
          "cfg": {
            "lr": 0.001,
            "epochs": 20
          },
          "mean": 17.79831099510193
        },
        {
          "cfg": {
            "lr": 0.003,
            "epochs": 20
          },
          "mean": 13.153547525405884
        },
        {
          "cfg": {
            "lr": 0.01,
            "epochs": 20
          },
          "mean": 7.754182696342468
        }
      ],
      "full": {
        "mean": 7.931941092014313,
        "std": 0.8101161551333251,
        "per_seed": [
          7.2100510597229,
          8.814193725585938,
          7.362153053283691,
          7.630332946777344,
          6.520580291748047,
          8.472823143005371,
          8.71557903289795,
          8.729815483093262
        ],
        "n": 8
      }
    },
    "idea": {
      "mean": 25.546241760253906,
      "std": 1.397921331795095,
      "per_seed": [
        26.91761589050293,
        25.963476181030273,
        26.16190528869629,
        22.51058578491211,
        25.61777687072754,
        26.621234893798828,
        24.128700256347656,
        26.448638916015625
      ],
      "n": 8,
      "best_cfg": {
        "lr": 0.01,
        "epochs": 20
      }
    },
    "comparison": {
      "delta_mean": 17.614300668239594,
      "idea_wins": 0,
      "n_pairs": 8,
      "p_value": 0.0081,
      "verdict": "idea worse (significant)",
      "system_worked": false
    },
    "mechanism_signature": {
      "sigma_predicted": 2.0,
      "slope_ratios_observed_trained_model": [
        0.49992549419403076,
        0.5001490116119385,
        0.5
      ],
      "median_ratio_observed": 0.5,
      "input_scale_2_output_abs_ratio": 1.0001696348190308,
      "confirmed": false
    }
  },
  "how_to_run": "/home/maxwelhelp/main/bin/python3 bench_run.py",
  "files": [
    "bench_run.py",
    "bench_report.json"
  ],
  "limitations": "Only the structurally matched tabular MLP track was tested. The activation used fixed sigma=2, eta=1, nu=0 and hard integer binning; no learned endpoint parameters, boundary blending, vision, sequence, or dynamics tracks were tested.",
  "system_verdict": "failed",
  "practical_verdict": "harms",
  "mechanism_ok": 0,
  "system_judged": true
}