| { | |
| "architectures": [ | |
| "QWenLMHeadModel" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration_qwen.QWenConfig", | |
| "AutoModelForCausalLM": "modeling_qwen.QWenLMHeadModel" | |
| }, | |
| "model_type": "qwen", | |
| "num_hidden_layers": 28, | |
| "num_attention_heads": 28, | |
| "num_key_value_heads": 4, | |
| "hidden_size": 4096, | |
| "intermediate_size": 11008, | |
| "hidden_act": "silu", | |
| "rotary_pct": 1.0, | |
| "rotary_emb_base": 10000, | |
| "max_position_embeddings": 131072, | |
| "initializer_range": 0.02, | |
| "rms_norm_eps": 1e-06, | |
| "use_cache": true, | |
| "vocab_size": 151936, | |
| "eos_token_id": 151643, | |
| "bos_token_id": 151643, | |
| "pad_token_id": 151643, | |
| "tie_word_embeddings": false, | |
| "use_parallel_residual": true, | |
| "bias": false, | |
| "rope_scaling": { | |
| "type": "linear", | |
| "factor": 4.0 | |
| }, | |
| "max_generate_tokens": 8192, | |
| "quantization_config": { | |
| "bits": 4, | |
| "group_size": 32, | |
| "quant_method": "gguf" | |
| }, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.32.0" | |
| } |