| { | |
| "model": "baichuan", | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 11008, | |
| "max_position_embeddings": 4096, | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "pad_token_id": 0, | |
| "layer_norm_eps": 1e-06, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "vocab_size": 125696, | |
| "segment_vocab_size": 0, | |
| "rope_rank": "updown", | |
| "skip_init": true, | |
| "norm_head": true, | |
| "generation_config": {"tokenizer_config": {"skip_special_tokens": true}, "max_length": 4096} | |
| } |