| { |
| "architectures": [ |
| "TinyLlavaForConditionalGeneration" |
| ], |
| "cache_dir": null, |
| "connector_type": "mlp2x_gelu", |
| "hidden_size": 896, |
| "ignore_index": -100, |
| "image_aspect_ratio": "square", |
| "image_token_index": -200, |
| "llm_model_name_or_path": "Qwen/Qwen2.5-0.5B", |
| "model_type": "tinyllava", |
| "num_queries": 128, |
| "num_resampler_layers": 3, |
| "pad_token": "<|endoftext|>", |
| "pad_token_id": 151643, |
| "resampler_hidden_size": 768, |
| "text_config": { |
| "_name_or_path": "Qwen/Qwen2.5-0.5B", |
| "architectures": [ |
| "Qwen2ForCausalLM" |
| ], |
| "bos_token_id": 151643, |
| "eos_token_id": 151643, |
| "hidden_size": 896, |
| "intermediate_size": 4864, |
| "max_position_embeddings": 32768, |
| "max_window_layers": 24, |
| "model_type": "qwen2", |
| "num_attention_heads": 14, |
| "num_hidden_layers": 24, |
| "num_key_value_heads": 2, |
| "rope_theta": 1000000.0, |
| "sliding_window": 32768, |
| "tie_word_embeddings": true, |
| "torch_dtype": "float16", |
| "use_mrope": false, |
| "use_sliding_window": false, |
| "vocab_size": 151936 |
| }, |
| "tokenizer_model_max_length": 2048, |
| "tokenizer_name_or_path": "Qwen/Qwen2.5-0.5B", |
| "tokenizer_padding_side": "right", |
| "tokenizer_use_fast": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.39.3", |
| "tune_type_connector": "full", |
| "tune_type_llm": "frozen", |
| "tune_type_vision_tower": "frozen", |
| "tune_vision_tower_from_layer": 0, |
| "use_cache": false, |
| "vision_config": { |
| "_name_or_path": "facebook/dinov2-small", |
| "apply_layernorm": true, |
| "architectures": [ |
| "Dinov2Model" |
| ], |
| "attention_probs_dropout_prob": 0.0, |
| "drop_path_rate": 0.0, |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.0, |
| "hidden_size": 384, |
| "image_size": 518, |
| "layer_norm_eps": 1e-06, |
| "layerscale_value": 1.0, |
| "mlp_ratio": 4, |
| "model_name_or_path": "facebook/dinov2-small", |
| "model_name_or_path2": "", |
| "model_type": "dinov2", |
| "num_attention_heads": 6, |
| "num_hidden_layers": 12, |
| "out_features": [ |
| "stage12" |
| ], |
| "out_indices": [ |
| 12 |
| ], |
| "patch_size": 14, |
| "qkv_bias": true, |
| "reshape_hidden_states": true, |
| "stage_names": [ |
| "stem", |
| "stage1", |
| "stage2", |
| "stage3", |
| "stage4", |
| "stage5", |
| "stage6", |
| "stage7", |
| "stage8", |
| "stage9", |
| "stage10", |
| "stage11", |
| "stage12" |
| ], |
| "torch_dtype": "float32", |
| "use_swiglu_ffn": false |
| }, |
| "vision_feature_layer": -2, |
| "vision_feature_select_strategy": "patch", |
| "vision_hidden_size": 384, |
| "vision_model_name_or_path": "facebook/dinov2-small", |
| "vision_model_name_or_path2": "", |
| "vocab_size": 151936 |
| } |
|
|
