How to use jeongyoonhuh/q_lora_korqa with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("google/gemma-3-4b-it") model = PeftModel.from_pretrained(base_model, "jeongyoonhuh/q_lora_korqa")
https://huggingface.co/jeongyoonhuh/q_lora_korqa/resolve/4db3b641cd1e81509ebc84aa3fcca5fcca151603/checkpoint-126/tokenizer_config.json
hf download hf://jeongyoonhuh/q_lora_korqa@4db3b641cd1e81509ebc84aa3fcca5fcca151603/checkpoint-126/tokenizer_config.json
curl -L -o tokenizer_config.json https://huggingface.co/jeongyoonhuh/q_lora_korqa/resolve/4db3b641cd1e81509ebc84aa3fcca5fcca151603/checkpoint-126/tokenizer_config.json
