{ "_name_or_path": "/data/spirin/llava/checkpoints/llavahf_gemma_lorarus_share/", "architectures": [ "LlavaForConditionalGeneration" ], "ignore_index": -100, "image_token_index": 256000, "model_type": "llava", "pad_token_id": 256001, "projector_hidden_act": "gelu", "text_config": { "_name_or_path": "google/gemma-2b-it", "architectures": [ "GemmaForCausalLM" ], "bos_token_id": 2, "eos_token_id": 1, "head_dim": 256, "hidden_act": "gelu", "hidden_activation": null, "hidden_size": 2048, "intermediate_size": 16384, "max_position_embeddings": 8192, "model_type": "gemma", "num_attention_heads": 8, "num_hidden_layers": 18, "num_key_value_heads": 1, "pad_token_id": 0, "rope_scaling": null, "tie_word_embeddings": true, "torch_dtype": "bfloat16", "vocab_size": 256064 }, "torch_dtype": "float32", "transformers_version": "4.42.4", "vision_config": { "hidden_size": 1024, "image_size": 336, "intermediate_size": 4096, "model_type": "clip_vision_model", "num_attention_heads": 16, "num_hidden_layers": 24, "patch_size": 14, "projection_dim": 768, "vocab_size": 32000 }, "vision_feature_layer": -2, "vision_feature_select_strategy": "default" }