New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

代码问题 #69

Open

ccjjs opened this issue Nov 14, 2024 · 0 comments

ccjjs commented Nov 14, 2024 •

edited

Loading

ALMA/utils/cpo_trainer.py

Line 606 in 7e0dae8

    
           penalty = torch.abs((policy_chosen_logps / num_non_pad_tokens[:len_chosen] - policy_rejected_logps / num_non_pad_tokens[len_chosen:]))

penalty = torch.clamp(self.relax_cofficient_1 * torch.exp(self.relax_cofficient_2 * penalty)-1, max=1.0)
这里penalty的计算用到了logits，并且没有detach，预期是penalty需要参与到梯度传播吗？

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment