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

(универсальный)

Ответ агента:

{
  "worked": false,
  "confidence": 9,
  "verdict": "Implemented moment-preserving rank-2 hidden-state compression as an end-to-end modification of the shared rnn_small architecture on the structurally matched dynamics track. Numerical and trained-model checks confirmed four hidden moments were preserved to about 1.4e-6 and 1.3e-6, but the idea had higher test MSE than the tuned baseline (0.007600 vs 0.006754), with paired delta +0.000846 and permutation p=0.4058; therefore it did not produce a significant win.",
  "metrics": {
    "baseline": "Dynamics rnn_small, 8 paired seeds, lr sweep {0.001, 0.003, 0.006}, best lr=0.006; mean test MSE 0.0067540, std 0.0028601.",
    "idea": "Moment-preserving rank-2 hidden compression, same 8 seeds and learning-rate union; mean test MSE 0.0075996, std 0.0019136. Paired delta +0.0008456, 3/8 wins, permutation p=0.4058.",
    "mechanism_signature": "Trained-model hidden-state moment residual 1.3113e-6; compression relative perturbation 0.4047; confirmed=true."
  },
  "how_to_run": "/home/maxwelhelp/main/bin/python3 bench_moment_preserving.py",
  "files": [
    "bench_moment_preserving.py",
    "bench_report.json"
  ],
  "limitations": "Only the built-in dynamics track was tested; no memory or runtime advantage was measured, and the implementation compresses each 64-dimensional GRU hidden state as an 8x8 matrix rather than a full HT tensor. The benchmark used 8 epochs and 400 training/400 test examples per seed.",
  "bench_report": {
    "bench_version": 1,
    "track": "dynamics",
    "model": "rnn_small",
    "metric_direction": "lower is better",
    "n_seeds": 8,
    "baseline": {
      "best_cfg": {
        "lr": 0.006
      },
      "sweep": [
        {
          "cfg": {
            "lr": 0.001
          },
          "mean": 0.021498297923244536
        },
        {
          "cfg": {
            "lr": 0.003
          },
          "mean": 0.0067647010437212884
        },
        {
          "cfg": {
            "lr": 0.006
          },
          "mean": 0.00675402011256665
        }
      ],
      "full": {
        "mean": 0.00675402011256665,
        "std": 0.0028601230170336904,
        "per_seed": [
          0.006943476852029562,
          0.005382216069847345,
          0.012206831015646458,
          0.006380172912031412,
          0.0068011293187737465,
          0.009766676463186741,
          0.0037104773800820112,
          0.0028411808889359236
        ],
        "n": 8
      }
    },
    "idea": {
      "mean": 0.007599645119626075,
      "std": 0.0019136134219505131,
      "per_seed": [
        0.009002679027616978,
        0.010305776260793209,
        0.008358007296919823,
        0.008188792504370213,
        0.00483678886666894,
        0.009082159027457237,
        0.006025956943631172,
        0.004997001029551029
      ],
      "n": 8
    },
    "comparison": {
      "delta_mean": 0.0008456250070594251,
      "idea_wins": 3,
      "n_pairs": 8,
      "per_seed_diffs": [
        0.0020592021755874157,
        0.004923560190945864,
        -0.003848823718726635,
        0.0018086195923388004,
        -0.001964340452104807,
        -0.0006845174357295036,
        0.002315479563549161,
        0.0021558201406151056
      ],
      "p_value": 0.4058,
      "mde": 0.002347660486662723,
      "mde_rel_pct": 34.75945359260367,
      "verdict": "no significant win",
      "system_worked": false
    },
    "mechanism_signature": {
      "math_check_max_moment_residual": 1.430511474609375e-06,
      "trained_model": {
        "raw_norm": 39.96965408325195,
        "compressed_relative_error": 0.40474483370780945,
        "moment_residual": 1.3113021850585938e-06,
        "target": "hidden moments",
        "predicted_residual": 0.0,
        "observed_residual": 1.3113021850585938e-06,
        "confirmed": true
      },
      "predicted": "four hidden-state moments preserved after rank-2 compression",
      "observed": "test-time hidden-state moment residual measured from trained model",
      "confirmed": true
    },
    "math_verification": {
      "max_moment_residual": 1.430511474609375e-06,
      "passed": true
    },
    "protocol": {
      "epochs": 8,
      "n_train": 400,
      "n_test": 400,
      "grid": [
        {
          "lr": 0.001
        },
        {
          "lr": 0.003
        },
        {
          "lr": 0.006
        }
      ],
      "seeds": [
        0,
        1,
        2,
        3,
        4,
        5,
        6,
        7
      ]
    }
  },
  "system_verdict": "partial",
  "practical_verdict": "inconclusive",
  "mechanism_ok": 1,
  "system_judged": true
}