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

Polish the error message and check for flash_attn. #58345

Merged
merged 13 commits into from
Dec 5, 2023

Conversation

Xreki
Copy link
Contributor

@Xreki Xreki commented Oct 24, 2023

PR types

Others

PR changes

OPs

Description

Pcard-70444
本PR的工作:

  • flashattn算子支持k、v不计算梯度。由于官方flash-attention库必须传入dk、dv的地址,故在k、v被设置成不计算梯度时,需要使用定义dk、dv的临时Tensor,并传入相应指针。
  • 一些代码上的轻微调整和日志上的优化。增加报错,不允许设置causal=True同时传入attn_mask

@paddle-bot
Copy link

paddle-bot bot commented Oct 24, 2023

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.

@paddle-bot paddle-bot bot added the contributor External developers label Oct 25, 2023
@paddle-bot paddle-bot bot removed the contributor External developers label Nov 3, 2023
Copy link

paddle-ci-bot bot commented Nov 14, 2023

Sorry to inform you that 45b0c33's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually.

Copy link
Contributor

@ZzSean ZzSean left a comment

Choose a reason for hiding this comment

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

LGTM for skipif

@Xreki Xreki requested a review from AnnaTrainingG December 4, 2023 06:28
@Xreki Xreki merged commit bf008c4 into PaddlePaddle:develop Dec 5, 2023
@Xreki Xreki deleted the op/polish_flashattn branch December 5, 2023 02:57
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