| { | |
| "_name_or_path": "Qwen/Qwen2-7B-Instruct", | |
| "architectures": [ | |
| "LlavaQwenSlowFastForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "cross_attn_every_n_layers": 8, | |
| "cross_attn_gating_type": "whole-dynamic-tanh-warmup", | |
| "cross_attn_implementation": "text-only-vanilla-flashattn", | |
| "cross_attn_max_layer_depth": 28, | |
| "cross_attn_min_layer_depth": 0, | |
| "eos_token_id": 151645, | |
| "fast_token_spatial_stride": 1, | |
| "fast_token_temporal_sampling_stride": 2, | |
| "fast_token_temporal_stride": 4, | |
| "hidden_act": "silu", | |
| "hidden_size": 3584, | |
| "image_aspect_ratio": "pad", | |
| "initializer_range": 0.02, | |
| "intermediate_size": 18944, | |
| "max_position_embeddings": 32768, | |
| "max_window_layers": 28, | |
| "min_fast_frames": 16, | |
| "mm_hidden_size": 3072, | |
| "mm_patch_merge_type": "flat", | |
| "mm_projector_lr": null, | |
| "mm_projector_type": "mlp2x_gelu", | |
| "mm_training_video_frames": 128, | |
| "mm_use_im_patch_token": false, | |
| "mm_use_im_start_end": false, | |
| "mm_video_pooling_stride": 2, | |
| "mm_vision_select_feature": "patch", | |
| "mm_vision_select_layer": -2, | |
| "mm_vision_tower": "convnext-576", | |
| "model_type": "llava_qwen_slow_fast", | |
| "num_attention_heads": 28, | |
| "num_hidden_layers": 28, | |
| "num_key_value_heads": 4, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000.0, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "tile_image_input": false, | |
| "tokenizer_model_max_length": 3072, | |
| "tokenizer_padding_side": "right", | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.45.1", | |
| "tune_mm_mlp_adapter": false, | |
| "use_cache": true, | |
| "use_mm_proj": true, | |
| "use_sliding_window": false, | |
| "vocab_size": 152064 | |
| } | |