{
  "architectures": ["GPT2LMHeadModel"],
  "hidden_size": 768,
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "vocab_size": 50257,
  "max_position_embeddings": 1024,
  "type_vocab_size": 2,
  "hidden_dropout_prob": 0.1,
  "attention_probs_dropout_prob": 0.1,
  "layer_norm_epsilon": 1e-5
}