joshcd/MNLP_M3_document_encoder
075
1{2 "architectures": [3 "Qwen2ForCausalLM"4 ],5 "attention_dropout": 0.0,6 "auto_map": {7 "AutoModel": "modeling_qwen.Qwen2Model",8 "AutoModelForCausalLM": "modeling_qwen.Qwen2ForCausalLM",9 "AutoModelForSequenceClassification": "modeling_qwen.Qwen2ForSequenceClassification"10 },11 "bos_token_id": 151643,12 "eos_token_id": 151643,13 "hidden_act": "silu",14 "hidden_size": 1536,15 "initializer_range": 0.02,16 "intermediate_size": 8960,17 "max_position_embeddings": 131072,18 "max_window_layers": 21,19 "model_type": "qwen2",20 "num_attention_heads": 12,21 "num_hidden_layers": 28,22 "num_key_value_heads": 2,23 "rms_norm_eps": 1e-06,24 "rope_theta": 1000000.0,25 "sliding_window": 131072,26 "tie_word_embeddings": false,27 "torch_dtype": "float32",28 "transformers_version": "4.41.2",29 "use_cache": true,30 "use_sliding_window": false,31 "is_causal": false,32 "vocab_size": 15164633}34 