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

【LLM Inference】Add Split MoE Op && Add Group MoE #69687

Open
wants to merge 19 commits into
base: develop
Choose a base branch
from

Conversation

gzy19990617
Copy link
Contributor

@gzy19990617 gzy19990617 commented Nov 25, 2024

PR Category

Inference

PR Types

Others

Description

card-71500

1.增加拆分的moe算子,更便于做精度对齐,将分组softmax操作合并到一个moe_dispatch算子之中,精度已对齐。

2.增加group_moe功能的支持,新增group_moe接口。

3.补充group moe下的拆分算子的精度单测

4.补充拆分算子与之前baseline不同精度的单测

Copy link

paddle-bot bot commented Nov 25, 2024

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@gzy19990617 gzy19990617 force-pushed the add_group_moe branch 4 times, most recently from 91d85bb to 316e922 Compare November 26, 2024 03:40
@gzy19990617 gzy19990617 changed the title 【LLM Inference】Add Split MOE Op && Add Group MOE 【LLM Inference】Add Split MoE Op && Add Group MoE Nov 26, 2024
@CLAassistant
Copy link

CLAassistant commented Dec 4, 2024

CLA assistant check
All committers have signed the CLA.

Copy link

paddle-ci-bot bot commented Dec 12, 2024

Sorry to inform you that f52f14b's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually.

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.

4 participants