[New Feature] fa3 支持flash mask #3184
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
gzy19990617
reviewed
Aug 5, 2025
| naive_attn_out = naive_attn(q_input, k_input, v_input, mask) | ||
| paddle_attn_out = paddle_flash_attn_mask(q_input, k_input, v_input, mask) | ||
|
|
||
| print((paddle_attn_out.reshape([-1]) - paddle.to_tensor(naive_attn_out).reshape([-1])).max()) |
yuanlehome
approved these changes
Aug 5, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
在flash attention 3的基础上支持了mask形式，其中mask是一个shape为[q_seq_len]的int 数组，对于第i个token，那么qkgemm矩阵的第i行，第mask[i]列之后的数字都会被mask 掉，即qkgemm[i,mask[i]:] 会被mask掉，例如要使用casual mask的话，mask的数组就为[1,2,3,4,5,.......]
若mask传入None，默认采用casual mask