{ "architectures": [ "LlavaForConditionalGeneration" ], "ignore_index": -100, "image_token_index": 32000, "model_type": "llava", "pad_token_id": 32001, "projector_hidden_act": "gelu", "text_config": { "_name_or_path": "TinyLlama/TinyLlama-1.1B-Chat-V1.0", "architectures": [ "LlamaForCausalLM" ], "hidden_size": 2048, "intermediate_size": 5632, "model_type": "llama", "num_hidden_layers": 22, "num_key_value_heads": 4, "rms_norm_eps": 1e-05, "torch_dtype": "bfloat16", "vocab_size": 32064 }, "torch_dtype": "float16", "transformers_version": "4.37.0.dev0", "vision_config": { "dropout": 0.0, "hidden_size": 1024, "image_size": 224, "intermediate_size": 4096, "model_type": "clip_vision_model", "num_attention_heads": 16, "num_hidden_layers": 24, "patch_size": 14, "projection_dim": 768 }, "vision_feature_layer": -2, "vision_feature_select_strategy": "default", "vocab_size": 32064 }