|
{ |
|
"_name_or_path": "../../huggingface/llava-onevision-qwen2-0.5b-ov/", |
|
"architectures": [ |
|
"LlavaQwenForCausalLM" |
|
], |
|
"attention_dropout": 0.0, |
|
"bert_base_path": null, |
|
"bos_token_id": 151643, |
|
"eos_token_id": 151645, |
|
"freeze_mm_mlp_adapter": false, |
|
"grounding_dino_config": null, |
|
"hidden_act": "silu", |
|
"hidden_size": 896, |
|
"image_aspect_ratio": "pad", |
|
"image_crop_resolution": null, |
|
"image_grid_pinpoints": null, |
|
"image_split_resolution": null, |
|
"image_token_index": 151646, |
|
"initializer_range": 0.02, |
|
"intermediate_size": 4864, |
|
"load_full_model": true, |
|
"load_ram": false, |
|
"max_position_embeddings": 32768, |
|
"max_window_layers": 24, |
|
"mm_hidden_size": 1152, |
|
"mm_newline_position": "grid", |
|
"mm_patch_merge_type": "flat", |
|
"mm_projector_lr": 2e-05, |
|
"mm_projector_type": "mlp2x_gelu", |
|
"mm_resampler_type": null, |
|
"mm_spatial_pool_mode": "bilinear", |
|
"mm_spatial_pool_stride": null, |
|
"mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model", |
|
"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": "../huggingface/siglip-so400m-patch14-384", |
|
"mm_vision_tower_lr": 2e-06, |
|
"model_type": "llava_qwen", |
|
"num_attention_heads": 14, |
|
"num_hidden_layers": 24, |
|
"num_key_value_heads": 2, |
|
"pos_skipping_range": 4096, |
|
"repeat_num": 3, |
|
"rms_norm_eps": 1e-06, |
|
"rope_scaling": null, |
|
"rope_theta": 1000000.0, |
|
"sliding_window": 32768, |
|
"tie_word_embeddings": true, |
|
"tokenizer_model_max_length": 2048, |
|
"tokenizer_padding_side": "right", |
|
"torch_dtype": "float16", |
|
"transformers_version": "4.37.2", |
|
"tune_mm_mlp_adapter": false, |
|
"use_cache": true, |
|
"use_mm_proj": true, |
|
"use_pos_skipping": false, |
|
"use_sliding_window": false, |
|
"vision_tower_pretrained": null, |
|
"vision_tower_weight_path": null, |
|
"vocab_size": 151647 |
|
} |
|
|