Enable Flash Attention #1065
Merged
Enable Flash Attention #1065
Conversation
This file contains 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
Signed-off-by: jiqing-feng <[email protected]>
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
jiqing-feng
changed the title
|
Hi @sywangyi , please review these changes. Thanks! |
sywangyi
reviewed
Jan 14, 2025
sywangyi
reviewed
Jan 14, 2025
sywangyi
reviewed
Jan 14, 2025
|
Hi @IlyasMoutawwakil. Do you mind reviewing this PR? Thanks! |
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
IlyasMoutawwakil
approved these changes
Jan 14, 2025
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
Signed-off-by: jiqing-feng <[email protected]>
|
It has 20%+ speed-up for 1st token over the patched models. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
In CPU torch 2.5 we got flash attention for prefill, it's faster than varlen attention. We also apply SDPA when no past_key_values found.
It also use fused linear on gpt2 model.