{ "_name_or_path": "/linzhihang/zhangyuhao/ACLlama_s2s/output/S2S/S2S_finetune_small_lr3e-5_magpie-slice-unit-languge-new-0717-filter_kd_offline_base_merge_0717", "adapter_size": 1280, "architectures": [ "ACLlamaForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "audio_token_len": 10, "audio_tower": "/linzhihang/LLMs/whisper-v3", "bos_token_id": 128000, "eos_token_id": [ 128001, 128008, 128009 ], "head_dim": 128, "hidden_act": "silu", "hidden_size": 3072, "initializer_range": 0.02, "intermediate_size": 8192, "max_position_embeddings": 131072, "mlp_bias": false, "model_type": "ACLlama", "num_attention_heads": 24, "num_hidden_layers": 28, "num_key_value_heads": 8, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": { "factor": 32.0, "high_freq_factor": 4.0, "low_freq_factor": 1.0, "original_max_position_embeddings": 8192, "rope_type": "llama3" }, "rope_theta": 500000.0, "tie_word_embeddings": true, "torch_dtype": "float16", "transformers_version": "4.48.1", "unit_output": "finetune_kd", "unit_vocab": "/linzhihang/zhangyuhao/unit_language/v5.8k.dict", "unit_vocab_size": 7801, "use_cache": false, "vocab_size": 128257 }