{
  "activation_function": "gelu_new",
  "attn_pdrop": 0.1,
  "average_logits": true,
  "bos_token_id": 0,
  "embd_pdrop": 0.1,
  "eos_token_id": 1,
  "head_locations": [
    8
  ],
  "head_weights": [
    0.7,
    0.3
  ],
  "initializer_range": 0.02,
  "layer_norm_epsilon": 1e-05,
  "model_type": "gpt2-multi-head",
  "n_embd": 512,
  "n_head": 8,
  "n_inner": null,
  "n_layer": 16,
  "n_positions": 256,
  "reorder_and_upcast_attn": false,
  "resid_pdrop": 0.1,
  "scale_attn_by_inverse_layer_idx": false,
  "scale_attn_weights": true,
  "summary_activation": null,
  "summary_first_dropout": 0.1,
  "summary_proj_to_labels": true,
  "summary_type": "cls_index",
  "summary_use_proj": true,
  "tie_additional_weights": false,
  "transformers_version": "4.31.0.dev0",
  "use_cache": true,
  "vocab_size": 500
}