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

[ZeRO-1] Sync features from r2.1_aws_neuron branch #7016

Closed
wants to merge 3 commits into from
Closed

[ZeRO-1] Sync features from r2.1_aws_neuron branch #7016

wants to merge 3 commits into from

Conversation

hgt312
Copy link
Collaborator

@hgt312 hgt312 commented May 1, 2024

Sync features from r2.1_aws_neuron branch:

  • use grad hook to make grad accumulation use higher precision
  • save master weights
  • an option to change cc ops' dtype

This PR also includes #6673

@hgt312
Copy link
Collaborator Author

hgt312 commented May 1, 2024

@jeffhataws

@jeffhataws jeffhataws self-requested a review May 1, 2024 23:26
@hgt312 hgt312 force-pushed the sync_features branch 2 times, most recently from 0a48474 to f9b969c Compare May 10, 2024 00:05
@hgt312 hgt312 mentioned this pull request May 29, 2024
Merged
@jeffhataws
Copy link
Collaborator

#7132 replaces this PR.

@hgt312 hgt312 closed this May 29, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jeffhataws jeffhataws Awaiting requested review from jeffhataws

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@hgt312 @jeffhataws