| { | |
| "activation_function": "gelu_new", | |
| "architectures": [ | |
| "GPTRefactForCausalLM" | |
| ], | |
| "attention_softmax_in_fp32": true, | |
| "attn_pdrop": 0.1, | |
| "bos_token_id": 50256, | |
| "do_sample": true, | |
| "embd_pdrop": 0.1, | |
| "eos_token_id": 50256, | |
| "initializer_range": 0.02, | |
| "layer_norm_epsilon": 1e-05, | |
| "model_type": "gpt_bigcode", | |
| "multi_query": true, | |
| "n_embd": 2048, | |
| "n_head": 32, | |
| "n_inner": null, | |
| "n_layer": 32, | |
| "n_positions": 1024, | |
| "resid_pdrop": 0.1, | |
| "scale_attention_softmax_in_fp32": true, | |
| "scale_attn_weights": true, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.28.1", | |
| "use_cache": true, | |
| "vocab_size": 49216 | |
| } | |