| { | |
| "architectures": [ | |
| "TinyLlavaForConditionalGeneration" | |
| ], | |
| "cache_dir": null, | |
| "connector_type": "mlp2x_gelu", | |
| "hidden_size": 576, | |
| "ignore_index": -100, | |
| "image_aspect_ratio": "square", | |
| "image_token_index": -200, | |
| "llm_model_name_or_path": "HuggingFaceTB/SmolLM2-135M-Instruct", | |
| "model_type": "tinyllava", | |
| "num_queries": 128, | |
| "num_resampler_layers": 3, | |
| "pad_token": "<|endoftext|>", | |
| "pad_token_id": 0, | |
| "resampler_hidden_size": 768, | |
| "text_config": { | |
| "_name_or_path": "HuggingFaceTB/SmolLM2-135M-Instruct", | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "hidden_size": 576, | |
| "initializer_range": 0.041666666666666664, | |
| "intermediate_size": 1536, | |
| "is_llama_config": true, | |
| "max_position_embeddings": 8192, | |
| "mlp_bias": false, | |
| "model_type": "llama", | |
| "num_attention_heads": 9, | |
| "num_hidden_layers": 30, | |
| "num_key_value_heads": 3, | |
| "pad_token_id": 2, | |
| "rms_norm_eps": 1e-05, | |
| "rope_interleaved": false, | |
| "rope_theta": 100000, | |
| "tie_word_embeddings": true, | |
| "torch_dtype": "float16", | |
| "transformers.js_config": { | |
| "kv_cache_dtype": { | |
| "fp16": "float16", | |
| "q4f16": "float16" | |
| } | |
| }, | |
| "vocab_size": 49152 | |
| }, | |
| "tokenizer_model_max_length": 2048, | |
| "tokenizer_name_or_path": "HuggingFaceTB/SmolLM2-135M-Instruct", | |
| "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": 49152 | |
| } | |