| { | |
| "activation_function": "gelu", | |
| "architectures": ["GPT3DevLMHeadModel"], | |
| "auto_map": { | |
| "AutoConfig": "modeling_gpt3dev.GPT3DevConfig", | |
| "AutoModel": "modeling_gpt3dev.GPT3DevModel", | |
| "AutoModelForCausalLM": "modeling_gpt3dev.GPT3DevLMHeadModel" | |
| }, | |
| "attn_pdrop": 0.0, | |
| "bos_token_id": 50256, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 50256, | |
| "initializer_range": 0.02, | |
| "layer_norm_epsilon": 1e-05, | |
| "model_type": "gpt3dev", | |
| "n_ctx": 2048, | |
| "n_embd": 768, | |
| "n_head": 12, | |
| "n_inner": 3072, | |
| "n_layer": 12, | |
| "n_positions": 2048, | |
| "reorder_and_upcast_attn": false, | |
| "resid_pdrop": 0.0, | |
| "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, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.46.1", | |
| "use_cache": true, | |
| "use_pre_layernorm": true, | |
| "vocab_size": 50257 | |
| } |