{
  "task": "reward_model",
  "strategy": "full",
  "base_model": "bert-base-uncased",
  "learning_rate": 2e-05,
  "epochs": 3,
  "n_train_pairs": 4000,
  "param_stats": {
    "total_params": 109483009,
    "trainable_params": 109483009,
    "trainable_pct": 100.0,
    "total_M": 109.48,
    "trainable_M": 109.483
  },
  "build_notes": [],
  "quantized": false,
  "val_history": [
    {
      "eval_loss": 0.006364240951370448,
      "eval_accuracy": 1.0,
      "eval_margin": 9.148771405220032,
      "eval_n": 500,
      "phase": "epoch_end",
      "strategy": "full",
      "epoch": 0
    },
    {
      "eval_loss": 0.0021165009238757193,
      "eval_accuracy": 1.0,
      "eval_margin": 10.256730794906616,
      "eval_n": 500,
      "phase": "epoch_end",
      "strategy": "full",
      "epoch": 1
    },
    {
      "eval_loss": 0.002454957519148593,
      "eval_accuracy": 1.0,
      "eval_margin": 10.375038027763367,
      "eval_n": 500,
      "phase": "epoch_end",
      "strategy": "full",
      "epoch": 2
    }
  ],
  "test_metrics": {
    "eval_loss": 0.0103483157624246,
    "eval_accuracy": 0.9983327775925308,
    "eval_margin": 10.615234009763027,
    "eval_n": 5998
  },
  "train_seconds": 336.67,
  "peak_gpu_mb": 5124.1,
  "data_source": "hf:nrizwan/safe_ai_assignment_1"
}