{
  "model_name": "Llama-10M-1M",
  "model_type": "causal-lm",
  "architecture": "LLaMA",
  "framework": "transformers",
  "created_at": "2025-07-05T16:11:06.857492",
  "parameters": {
    "total": 3652032,
    "hidden_size": 192,
    "num_layers": 6,
    "num_heads": 6,
    "vocab_size": 288,
    "sequence_length": 128
  },
  "training": {
    "dataset_size": 3519,
    "epochs": 2,
    "batch_size": 32,
    "learning_rate": "3e-4",
    "training_time_seconds": 26.0941,
    "final_loss": 2.5500883795998313
  },
  "evaluation": {
    "perplexity": 33.73914194244188,
    "diversity_score": 0.9286521711438311,
    "repetition_score": 0.5445736434108527,
    "top_token_confidence": 0.3424334205046762,
    "entropy": 2.0452219695628933,
    "num_eval_samples": 100
  }
}