Skip to content

[WIP] Use DTensor-based tensor parallel #180

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
Open
wants to merge 1 commit into
base: gh/kwen2501/1/base
Choose a base branch
from
Open

[WIP] Use DTensor-based tensor parallel #180

wants to merge 1 commit into from

Conversation

kwen2501
Copy link

@kwen2501 kwen2501 commented Jun 12, 2024 •
edited
Loading

Stack from ghstack (oldest at bottom):

Status:

  • Switched to DTensor based TP in regular tensor path
  • Result is correct, but there is a perf gap (seems to perform extra colls in the beginning, investigating)
  • TODO: switch to DTensor for quantized path too

kwen2501 added a commit that referenced this pull request Jun 12, 2024
@kwen2501
Use DTensor-based tensor parallel
eaae2f5 
ghstack-source-id: b55b264
Pull Request resolved: #180
@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jun 12, 2024
@kwen2501 kwen2501 changed the title Use DTensor-based tensor parallel [WIP] Use DTensor-based tensor parallel Jun 12, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@kwen2501 @facebook-github-bot