{
  "architectures": [
    "GPT2LMHeadModel"
  ],
  "model_type": "gpt2",
  "n_embd": 768,
  "n_layer": 12,
  "n_head": 12,
  "vocab_size": 50257,
  "n_positions": 1024,
  "n_ctx": 1024,
  "activation_function": "gelu",
  "initializer_range": 0.02,
  "layer_norm_epsilon": 1e-05,
  "attn_pdrop": 0.1,
  "embd_pdrop": 0.1,
  "resid_pdrop": 0.1,
  "summary_activation": null,
  "summary_proj_to_labels": true,
  "summary_first_dropout": 0.1,
  "gradient_checkpointing": false,
  "use_cache": true,
  "scale_attn_weights": true,
  "reorder_and_upcast_attn": false
}