rlhf.py: error: unrecognized arguments GRPO训练无法启动

[hjx-bupt]

Describe the bug
What the bug is, and how to reproduce, better with screenshots(描述bug以及复现过程，最好有截图)

Your hardware and system info
Write your system info like CUDA version/system/GPU/torch version here(在这里给出硬件信息和系统信息，如CUDA版本，系统，GPU型号和torch版本等)
CUDA 12.9
VLLM 0.8.5POST1
TRANSFORMERS 4.51.1

Additional context
Add any other context about the problem here(在这里补充其他信息)
很迷，有遇到过的同学吗