Skip to content
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

model.float() OOM as #3510 #3606

Closed
1 task done
lk137095576 opened this issue May 7, 2024 · 2 comments
Closed
1 task done

model.float() OOM as #3510 #3606

lk137095576 opened this issue May 7, 2024 · 2 comments
Labels
solved This problem has been already solved

Comments

@lk137095576
Copy link

Reminder

  • I have read the README and searched the existing issues.

Reproduction

#3510
看下面的图，反向是用的fp16的，计算之后才后cast为32的。
从下面微软介绍deepspeed的视频也可以看到，反向的时候，用的也是16。
https://www.microsoft.com/en-us/research/blog/zero-deepspeed-new-system-optimizations-enable-training-models-with-over-100-billion-parameters/

这里提到过，但是我不知道如何reopen issue，所以新提了一个。

Expected behavior

No response

System Info

No response

Others

No response
@hiyouga hiyouga added the pending This problem is yet to be addressed label May 7, 2024
@lk137095576
Copy link
Author

我今天又看了下，我测试的是14B，如果是72B full的话，即使是80G的显卡也撑不过model.float()

@hiyouga hiyouga added the enhancement New feature or request label May 15, 2024
@hiyouga
Copy link
Owner

hiyouga commented May 15, 2024

fixed

@hiyouga hiyouga added solved This problem has been already solved and removed enhancement New feature or request pending This problem is yet to be addressed labels May 15, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
solved This problem has been already solved
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@hiyouga @lk137095576