Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[XPU] Llama XPU's swiglu uses phi's swiglu #9414

Merged
merged 1 commit into from
Nov 13, 2024

Conversation

dynamicheart
Copy link
Contributor

@dynamicheart dynamicheart commented Nov 12, 2024

PR types

Function optimization

PR changes

Models

Description

Llama swiglu XPU调用逻辑使用Paddle phi原生的swiglu算子

Llama2 13B tp2 20steps回归测试

调用paddle_xpu_nn.xpu_swiglu的Loss
image

调用paddle.incubate.nn.functional.swiglu的Loss
image

Copy link

paddle-bot bot commented Nov 12, 2024

Thanks for your contribution!

@paddle-bot paddle-bot bot added the XPU label Nov 12, 2024
@dynamicheart dynamicheart changed the title [XPU] Llama XPU's swiglu uses Paddle's native swiglu [XPU] Llama XPU's swiglu uses phi's swiglu Nov 12, 2024
Copy link
Collaborator

@wawltor wawltor left a comment

Choose a reason for hiding this comment

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

LGTM

@wawltor wawltor merged commit 2838e80 into PaddlePaddle:develop Nov 13, 2024
9 of 12 checks passed
# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants