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

v100下无法使用int8训练 #406

Closed
lucasjinreal opened this issue May 25, 2023 · 3 comments
Closed

v100下无法使用int8训练 #406

lucasjinreal opened this issue May 25, 2023 · 3 comments

Comments

@lucasjinreal
Copy link

load_in_8bit 在v100 下非常慢，但是好像用float16 会loss 起飞，用float32 会OOM，怎么弄
@uuser0748
Copy link

v100就是不支持int8。
Facico/Chinese-Vicuna#39
参考这个

@xianghuisun
Copy link
Collaborator

https://github.com/LianjiaTech/BELLE/blob/main/train/docs/FAQ.md
可参考FAQ中 关于V100机器上8bit量化训练失败或loss不正常

@xianghuisun
Copy link
Collaborator

load_in_8bit 在v100 下非常慢，但是好像用float16 会loss 起飞，用float32 会OOM，怎么弄

您如果是多卡V100，也可采用stage3，应该也能训起来

@zhangzhenyu13 zhangzhenyu13 mentioned this issue May 29, 2023
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@lucasjinreal @xianghuisun @uuser0748