-
Notifications
You must be signed in to change notification settings - Fork 1.6k
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
Adalora deepspeed #1625
Adalora deepspeed #1625
Conversation
d297a96
to
6ed60f7
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks a lot, these changes make sense. I have some minor comments but no real blockers. Also pinging @pacman100 as our DS expert.
6ed60f7
to
68fac81
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you @sywangyi for the fixes when using AdaLoRA with DeepSpeed Stage 3 😄! Left a comment inline with a suggestion from Benjamin.
Signed-off-by: Wang, Yi A <[email protected]>
68fac81
to
513ca8d
Compare
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the adjustments, LGTM. Sourab, please merge when you think it's ready.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you @sywangyi for iterating and simplifying the logic!
Signed-off-by: Wang, Yi A <[email protected]>
Signed-off-by: Wang, Yi A <[email protected]>
fix crash when enable deepspeed zero3 + adalora