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

Init on the xla device #521

Merged
merged 5 commits into from
Apr 3, 2024
Merged

Init on the xla device #521

merged 5 commits into from
Apr 3, 2024

Conversation

michaelbenayoun
Copy link
Member

@michaelbenayoun michaelbenayoun commented Mar 20, 2024

What does this PR do?

  • Initializes parallel layers on the xla device directly, instead of the cpu device. It saves a lot of host memory.
  • It does the same for the weights that are initialized during lazy load.

@HuggingFaceDocBuilderDev

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.

@michaelbenayoun michaelbenayoun marked this pull request as ready for review April 2, 2024 14:49
Copy link
Collaborator

@dacorvo dacorvo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thanks !

@michaelbenayoun michaelbenayoun merged commit 12b06a3 into main Apr 3, 2024
8 of 12 checks passed
@michaelbenayoun michaelbenayoun deleted the init_on_xla branch April 3, 2024 12:47
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

Successfully merging this pull request may close these issues.

3 participants