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.

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

DP FSDP & SP #61

Open
wangpengfei1013 opened this issue Mar 14, 2024 · 0 comments
Open

DP FSDP & SP #61

wangpengfei1013 opened this issue Mar 14, 2024 · 0 comments

Comments

@wangpengfei1013
Copy link

How should we understand data parallelism and fully sharded data parallelism?They should be a definition in terms of the previous understanding?
Another question is that SP and TP constitute MP groups. Does it mean that as many SP groups as many complete model copies are needed?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant