| { | |
| "_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct", | |
| "architectures": [ | |
| "LlavaQwen2ForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "eos_token_id": 151645, | |
| "freeze_mm_mlp_adapter": false, | |
| "hidden_act": "silu", | |
| "hidden_size": 1536, | |
| "image_aspect_ratio": "pad", | |
| "initializer_range": 0.02, | |
| "intermediate_size": 8960, | |
| "max_position_embeddings": 32768, | |
| "max_window_layers": 21, | |
| "mm_hidden_size": 3072, | |
| "mm_patch_merge_type": "flat", | |
| "mm_projector_lr": null, | |
| "mm_projector_type": "mlp2x_gelu", | |
| "mm_use_im_patch_token": false, | |
| "mm_use_im_start_end": false, | |
| "mm_vision_select_feature": "patch", | |
| "mm_vision_select_layer": -2, | |
| "mm_vision_tower": "laion/CLIP-convnext_xxlarge-laion2B-s34B-b82K-augreg-soup", | |
| "model_type": "llava_qwen2", | |
| "num_attention_heads": 12, | |
| "num_hidden_layers": 28, | |
| "num_key_value_heads": 2, | |
| "only_tune_vision_tower": false, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000.0, | |
| "sliding_window": null, | |
| "tie_word_embeddings": true, | |
| "tokenizer_model_max_length": 1024, | |
| "tokenizer_padding_side": "right", | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.45.1", | |
| "tune_mm_mlp_adapter": false, | |
| "tune_vision_tower": false, | |
| "use_cache": true, | |
| "use_mm_proj": true, | |
| "use_sliding_window": false, | |
| "vocab_size": 151936 | |
| } | |