lusongshuo-mt

Results 1 issues of lusongshuo-mt

运行ppo_ray训练qwen2 72B的时候经常会报错 ![image](https://github.com/user-attachments/assets/b55ab8cc-c8fa-40ba-8aa2-4bed3938e756) 运行脚本关键参数如下,已使用官方推荐decker: ray job submit --address="http://127.0.0.1:8265" \ --runtime-env-json='{"working_dir": "/openrlhf", "pip": "/openrlhf/requirements.txt"}' \ -- python3 examples/train_ppo_ray.py \ --ref_num_nodes 2 \ --ref_num_gpus_per_node 8 \ --reward_num_nodes 1 \ --reward_num_gpus_per_node 8 \...