| { | |
| "_name_or_path": "/cpfs02/shared/llmit6/liudawei/xpuyu_work_dirs/internlm2-1_8b-myds-llada-v3/hf-310000-anneal4k", | |
| "architectures": [ | |
| "InternLM2ForCausalLM" | |
| ], | |
| "attn_implementation": "sdpa", | |
| "auto_map": { | |
| "AutoConfig": "configuration_internlm2.InternLM2Config", | |
| "AutoModel": "modeling_internlm2.InternLM2ForCausalLM", | |
| "AutoModelForCausalLM": "modeling_internlm2.InternLM2ForCausalLM" | |
| }, | |
| "bias": false, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "fuse_cross_entropy": true, | |
| "hidden_act": "silu", | |
| "hidden_size": 2048, | |
| "initializer_range": 0.006, | |
| "intermediate_size": 8192, | |
| "is_causal": false, | |
| "max_position_embeddings": 32768, | |
| "model_type": "internlm2", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 24, | |
| "num_key_value_heads": 8, | |
| "pad_token_id": 2, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.46.0", | |
| "use_cache": false, | |
| "vocab_size": 128512 | |
| } | |