{
  "architectures": [
    "LlamaForCausalLM"
  ],
  "model_type": "llama",
  "hidden_size": 5120,
  "intermediate_size": 13824,
  "num_hidden_layers": 1,
  "num_attention_heads": 40,
  "num_key_value_heads": 8,
  "head_dim": 128,
  "vocab_size": 152064,
  "draft_vocab_size": 32000,
  "rope_theta": 1000000.0,
  "rms_norm_eps": 1e-05,
  "torch_dtype": "bfloat16",
  "tie_word_embeddings": false
}