Visual Question Answering
Transformers
Safetensors
llava
image-text-to-text
AIGC
LLaVA
Inference Endpoints
Human_LLaVA / config.json
zhangYuanHui's picture
update
babae27
{
"architectures": [
"LlavaForConditionalGeneration"
],
"ignore_index": -100,
"image_token_index": 128257,
"model_type": "llava",
"pad_token_id": 128258,
"projector_hidden_act": "gelu",
"text_config": {
"_name_or_path": "./iter_45000_ft/llm_merge_lora",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 128000,
"eos_token_id": 128009,
"intermediate_size": 14336,
"max_position_embeddings": 8192,
"model_type": "llama",
"num_key_value_heads": 8,
"rms_norm_eps": 1e-05,
"rope_theta": 500000.0,
"torch_dtype": "float16",
"vocab_size": 128320
},
"torch_dtype": "float16",
"transformers_version": "4.43.3",
"vision_config": {
"_name_or_path": "./iter_45000_ft/vit_merge_lora",
"architectures": [
"SiglipVisionModel"
],
"hidden_act": "gelu_pytorch_tanh",
"hidden_size": 1152,
"image_size": 384,
"intermediate_size": 4304,
"layer_norm_eps": 1e-06,
"model_type": "siglip_vision_model",
"num_attention_heads": 16,
"num_hidden_layers": 27,
"patch_size": 14,
"torch_dtype": "float32"
},
"vision_feature_layer": -2,
"vision_feature_select_strategy": "default"
}