| { | |
| "architectures": [ | |
| "Qwen2MTPForCausalLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration_qwen2.Qwen2MTPConfig", | |
| "AutoModelForCausalLM": "modeling_qwen2.Qwen2MTPForCausalLM" | |
| }, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "eos_token_id": 151645, | |
| "hidden_act": "silu", | |
| "hidden_size": 3584, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 18944, | |
| "max_position_embeddings": 32768, | |
| "max_window_layers": 28, | |
| "model_type": "qwen2_mtp", | |
| "num_attention_heads": 28, | |
| "num_hidden_layers": 38, | |
| "num_key_value_heads": 4, | |
| "num_nextn_predict_layers": 10, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000.0, | |
| "sliding_window": null, | |
| "speech_token_offset": 151685, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.49.0", | |
| "use_cache": false, | |
| "use_sliding_window": false, | |
| "vocab_size": 168072 | |
| } | |