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

(универсальный)

Ответ агента:

{
  "worked": false,
  "confidence": 9,
  "verdict": "Built a tied-GRU refinement system on the structurally matched dynamics track, comparing fixed-cap inference against spectral-gap adaptive halting with identical independently trained architectures, data, epochs, and shared hyperparameter settings. Across 8 paired seeds, the best idea and baseline had identical MSE (0.0043454), paired delta 0 and permutation p=1.0; the mechanism signature was also not confirmed because predicted tau did not correlate with observed iterations.",
  "metrics": {
    "baseline": "Best config lr=0.01, cap=4; full 8-seed mean test MSE=0.004345412977272645.",
    "idea": "Best config lr=0.01, cap=4, rtol=0.025, tau=3.0; full 8-seed mean test MSE=0.004345412977272645; paired delta=0.0, p=1.0; mean predicted tau=7.15865, mean observed iterations=4.0, correlation=0.0, confirmed=false."
  },
  "how_to_run": "/home/maxwelhelp/main/bin/python3 spectral_bench.py",
  "files": [
    "spectral_bench.py",
    "bench_report.json",
    "full_run.log"
  ],
  "limitations": "This tested the intervention on the built-in GRU dynamics model rather than a looped transformer. Wall-clock/JVP overhead was not separately benchmarked, the controller used a finite-difference directional JVP rather than an exact autodiff power iteration, and the adaptive rule did not reduce the selected cap on this task. The stage-2 result is therefore a no-effect outcome, not evidence of an inference-speed win.",
  "bench_report": {
    "bench_version": 1,
    "track": "dynamics",
    "model": "rnn_small",
    "metric_direction": "lower is better",
    "n_seeds": 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
    },
    "mechanism_signature": {
      "quantity": "trained-model predicted local tau vs observed adaptive iterations",
      "n_examples": 6400,
      "predicted_tau_mean": 7.158651609644294,
      "observed_iterations_mean": 4.0,
      "correlation": 0.0,
      "confirmed": false
    },
    "baseline": {
      "best_cfg": {
        "lr": 0.01,
        "cap": 4,
        "rtol": 0.025,
        "tau": 3.0
      },
      "sweep_configs": 9,
      "full_mean": 0.004345412977272645
    },
    "idea": {
      "cfg": {
        "lr": 0.01,
        "cap": 4,
        "rtol": 0.025,
        "tau": 3.0
      },
      "mean": 0.004345412977272645,
      "alternatives": [
        {
          "lr": 0.01,
          "cap": 4,
          "rtol": 0.025,
          "tau": 3.0
        },
        {
          "lr": 0.001,
          "cap": 4,
          "rtol": 0.025,
          "tau": 3.0
        },
        {
          "lr": 0.01,
          "cap": 3,
          "rtol": 0.025,
          "tau": 3.0
        }
      ]
    }
  },
  "system_verdict": "failed",
  "practical_verdict": "no_effect",
  "mechanism_ok": 0,
  "system_judged": true
}