{
  "vocab_size": 8000,
  "block_size": 384,
  "n_layer": 8,
  "n_head": 7,
  "n_embd": 448,
  "dropout": 0.2,
  "tokenizer_type": "sentencepiece_bpe",
  "tokenizer_file": "MTP1_tokenizer.model",
  "special_tokens": {
    "pad_id": 0,
    "bos_id": 1,
    "eos_id": 2,
    "unk_id": 3
  },
  "generation_defaults": {
    "max_new_tokens": 200,
    "temperature": 1.05,
    "top_k": 40,
    "top_p": 0.95,
    "repetition_penalty": 1.2,
    "no_repeat_ngram_size": 3
  },
  "meta": {
    "model_name": "MTP-1",
    "architecture": "gpt_rope_swiglu_rmsnorm",
    "trained_examples": 18751,
    "augmented_examples": 54208,
    "corpus_fragments": 573,
    "total_sequences": 56651,
    "val_sequences": 1927,
    "best_val_loss": 4.256011638641358,
    "stopped_early": false,
    "label_smoothing": 0.1,
    "created": "2026-08-09 15:48:32"
  }
}