SDPA and memory saving tricks #23
Merged
+286
−37
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
|
Closes #18 lol nvm that only works for issues |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #23 +/- ##
==========================================
+ Coverage 54.47% 55.64% +1.17%
==========================================
Files 7 8 +1
Lines 380 487 +107
==========================================
+ Hits 207 271 +64
- Misses 173 216 +43 ☔ View full report in Codecov by Sentry. |
Fixing key chunk when sampling
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Outdated suggestions cannot be applied.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.
We either need to add a function that swaps the sdpa out for all layers then swapped at the top of the transformer, or, even better, make the sdpa come in with the model call instead of being a field in the layers.