HA-DPO
Collection
Collections for Hallucination-aware Direct Preference Optimization • 7 items • Updated
How to use juliozhao/hadpo-llava-1.5 with PEFT:
from peft import PeftModel
from transformers import AutoModelForCausalLM
base_model = AutoModelForCausalLM.from_pretrained("liuhaotian/llava-v1.5-7b")
model = PeftModel.from_pretrained(base_model, "juliozhao/hadpo-llava-1.5")