Thanks to visit codestin.com
Credit goes to github.com

Skip to content

16-bit float reductions + updated softmax #399

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
Merged
merged 1 commit into from
Apr 1, 2023
Merged

16-bit float reductions + updated softmax #399

merged 1 commit into from
Apr 1, 2023

Conversation

@cliffburdick
Copy link
Collaborator

@cliffburdick cliffburdick commented Mar 29, 2023 •
edited
Loading

  • Added 16-bit support for several reduction types (both __half and __nvbfloat16
  • Added unit tests for 16-bit reductions supported
  • Added 16-bit support to softmax
  • Added bias term on softmax to match Python's implementation

@cliffburdick
Copy link
Collaborator Author

/blossom-ci

tylera-nvidia
tylera-nvidia approved these changes Mar 31, 2023
View reviewed changes
@cliffburdick
Copy link
Collaborator Author

/blossom-ci

@cliffburdick cliffburdick merged commit c9a2521 into main Apr 1, 2023
@cliffburdick cliffburdick deleted the softmax2 branch April 1, 2023 03:49
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@luitjens luitjens luitjens approved these changes

@tylera-nvidia tylera-nvidia tylera-nvidia approved these changes

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@cliffburdick @luitjens @tylera-nvidia