| { | |
| "_name_or_path": "PIIP-LLaVA_ConvNeXt-B_CLIP-L_640-224_7B", | |
| "architectures": [ | |
| "LlavaLlamaForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "freeze_mm_mlp_adapter": false, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "image_aspect_ratio": "pad", | |
| "image_feature_multiply": 1.0, | |
| "image_grid_pinpoints": null, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 11008, | |
| "max_position_embeddings": 4096, | |
| "mm_hidden_size": 1024, | |
| "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": "configs/piip-llava_convnext-b_clip-l_640-224_7B.py", | |
| "model_type": "llava_llama", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 32, | |
| "pad_token_id": 0, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "tokenizer_model_max_length": 2048, | |
| "tokenizer_padding_side": "right", | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.37.2", | |
| "tune_mm_mlp_adapter": false, | |
| "tune_vision_tower": true, | |
| "tune_vit_pos_embedding": false, | |
| "use_cache": true, | |
| "use_mm_proj": true, | |
| "vision_tower_name": "configs/piip-llava_convnext-b_clip-l_640-224_7B.py", | |
| "vocab_size": 32000 | |
| } | |