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

FP16 微调overflow #43

Closed

mynewstart opened this issue Jul 13, 2023 · 1 comment

mynewstart commented Jul 13, 2023

请问有人试过用DeepSpeed Chat的代码基于ZERO2+FP16微调吗，发现会下溢出, 换成BF16后同样发现embeddng 层某些参数变为0，导致后面梯度为nan，没法更新参数，请问怎么解决？

Contributor

GradientGuru commented Jul 18, 2023

modeling代码已经更新，请在hugging face pull 最新代码。微调可参考本项目README更新

GradientGuru closed this as completed

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