| { | |
| "architectures": [ | |
| "GPTJiangForCausalLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration_gpt_jiang.GPTJiangConfig", | |
| "AutoModel": "modeling_gpt_jiang.GPTJiangModel", | |
| "AutoModelForCausalLM": "modeling_gpt_jiang.GPTJiangForCausalLM" | |
| }, | |
| "bos_token_id": 0, | |
| "eos_token_id": 0, | |
| "gated": true, | |
| "hidden_act": "gelu_pytorch_tanh", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 12288, | |
| "layer_norm_eps": 1e-05, | |
| "max_position_embeddings": 4096, | |
| "mlp_bias": false, | |
| "model_type": "gpt_jiang", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 48, | |
| "rotary_emb_base": 10000, | |
| "rotary_pct": 1.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.30.2", | |
| "use_cache": false, | |
| "use_parallel_residual": true, | |
| "vocab_size": 57000 | |
| } | |