Thank you for your great work.

I followed the recommended setup exactly, but the hidden state is returning NaN, and I'm not sure why. I am using:

model_path = "liuhaotian/llava-v1.5-7b"
model_name = get_model_name_from_path(model_path)
tokenizer, model, image_processor, context_len = load_pretrained_model(
model_path=model_path, model_base=None, model_name=model_name,
load_4bit=False, load_8bit=False
)

My prompt is:

A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER:

Hint: Please answer the question and provide the final answer at the end.

Question: What is the total number of objects in the image? ASSISTANT: