fix: ExLlama Backend Context Size & Rope Scaling #1311
Merged
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
lunamidori5
approved these changes
Nov 21, 2023
There was a problem hiding this comment.
@mudler why does it keep asking me for a review everytime I add you lol
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Outdated suggestions cannot be applied.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.
Description
This PR fixes ExLlama backend having fixed 2048 size cache. This results in the following error when 2048 context is exceeded:
stderr RuntimeError: start (2009) + length (44) exceeds dimension size (2048)This PR also passes through Rope scaling config to the ExLlama backend.
Notes for Reviewers
These attributes are set by default here, and not loaded from the model config json. Therefore, we need to set it from LocalAI's end. It would also be nice to add warnings for all backends of unimplemented config flags that the user defined in the yaml.
Signed commits