-
Notifications
You must be signed in to change notification settings - Fork 30
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
chore: Remove flash-attention v1 (#38)
Removing flash-attention v1 from the server-release image, to speed up build times. Modifications: - remove flash-att-v1 build stage from Dockerfile - remove server/Makefile-flash-att - create GitHub package for cache image on GitHub container registry - push full cache image to ghcr,io on push to main (PR merged) - use cache image from ghcr.io for PR builds - replace build stages/step with single `build-and-push` action - temporarily build dropout_layer_norm and rotary_emb from flash-attention v2 --------- Signed-off-by: Christian Kadner <[email protected]>
- Loading branch information
Showing
6 changed files
with
168 additions
and
181 deletions.
There are no files selected for viewing
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
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,2 +1,6 @@ | ||
# exclude any files inside the .git folder to not invalidate docker layer caches | ||
.git | ||
|
||
# exclude build artifacts | ||
target | ||
server/transformers | ||
server/transformers |
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
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
Oops, something went wrong.