How to use LARK-Lab/SWITCH-Phase3-GRPO-LoRA-Qwen3-8B with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("/root/.cache/modelscope/hub/models/Qwen/Qwen3-8B") model = PeftModel.from_pretrained(base_model, "LARK-Lab/SWITCH-Phase3-GRPO-LoRA-Qwen3-8B")
