Jsjgjhg comments

Results 5 comments of


                                            Jsjgjhg

Did you solve this problem?

> Is your question about the metrics to use? Yes，how to measure whether a reward model is beneficial for PPO training?

Why is indicator ppo/policy/loss always negative?like this: ![Uploading 1726630796123.png…]()

I hope to receive your guidance, thank you

Thank you.