config.json

{
    "_name_or_path": "gpt2",
    "architectures": [
      "GPT2LMHeadModel"
    ],
    "attention_probs_dropout_prob": 0.1,
    "bos_token_id": 50256,
    "eos_token_id": 50256,
    "initializer_range": 0.02,
    "layer_norm_eps": 1e-5,
    "max_position_embeddings": 1024,
    "model_type": "gpt2",
    "n_positions": 1024,
    "n_ctx": 1024,
    "n_embd": 768,
    "n_layer": 12,
    "n_head": 12,
    "vocab_size": 50257,
    "transformers_version": "4.45.1",
    "use_cache": true,
    "pad_token_id": 50256,
    "type_vocab_size": 1,
    "hidden_dropout_prob": 0.1,
    "activation_function": "gelu_new",
    "gradient_checkpointing": false,
    "tokenizer_class": "GPT2Tokenizer",
    "model_parallel": false
  }