Elide logits if the logits pointer parameter is NULL #107
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
saharNooby
reviewed
Jun 26, 2023
|
Please add info about possibility of skipping logits calculation into documentation in I would also suggest implementing it in PyTorch wrapper; but this is optional, especially given that the wrapper will be rewritten soon. |
This speeds up sequence mode evaluations by up to 20% if you ingest a large prompt and then only retrieve the logits at the very end. Note that you must pass a NULL pointer to the logits parameter in order to take advantage of this optimization.
LoganDark
force-pushed
the
logits-elision
branch
from
185243a
to
f5d6af9
Compare
saharNooby
approved these changes
Jun 27, 2023
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Outdated suggestions cannot be applied.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.
There is no need to calculate logits when it is not necessary, so perform some trickery to avoid calculating that part of the cgraph if logits is not actually going to be used.
Depends on #103
Discussed on #106 (comment)