| { | |
| "_name_or_path": "/data4/liuyang/checkpoints/xl_g_line_s2_decay_exp10_260k_sft_v2_dedup/iter-0020000-minicpm", | |
| "architectures": [ | |
| "MiniCPMForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_minicpm.MiniCPMConfig", | |
| "AutoModel": "modeling_minicpm.MiniCPMModel", | |
| "AutoModelForCausalLM": "modeling_minicpm.MiniCPMForCausalLM", | |
| "AutoModelForSeq2SeqLM": "modeling_minicpm.MiniCPMForCausalLM", | |
| "AutoModelForSequenceClassification": "modeling_minicpm.MiniCPMForSequenceClassification" | |
| }, | |
| "bos_token_id": 1, | |
| "dim_model_base": 256, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_act_param": 0.03, | |
| "hidden_size": 1536, | |
| "initializer_range": 0.1, | |
| "intermediate_size": 3840, | |
| "max_position_embeddings": 131072, | |
| "mlp_bias": false, | |
| "model_type": "minicpm", | |
| "num_attention_heads": 24, | |
| "num_hidden_layers": 48, | |
| "num_key_value_heads": 8, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 500000.0, | |
| "scale_depth": 1.4, | |
| "scale_emb": 12, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.49.0", | |
| "use_cache": true, | |
| "vocab_size": 65280 | |
| } | |