{
  "architectures": [
    "GPTForHF"
  ],
  "bias": false,
  "block_size": 1024,
  "dropout": 0.0,
  "model_type": "gpt",
  "n_embd": 768,
  "n_head": 12,
  "n_layer": 12,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.44.1",
  "vocab_size": 50258
}