Fixes #2087 - _process_tokens for empty prompts in KTOTrainer #2093
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
The function _process_tokens in trl/trainers/kto_trainer.py crashes if the prompt_input_ids are an empty list. - added a check for nonzero length - added a check for nonzero length of answer_input_ids for consistency The checks happen when determining when subtracting 1 from max_length (happens when BOS or EOS is already present).
kashif
approved these changes
Sep 21, 2024
|
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. |
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Outdated suggestions cannot be applied.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.
What does this PR do?
The function _process_tokens in trl/trainers/kto_trainer.py crashes if the prompt_input_ids are an empty list.
The checks happen when determining if 1 should be subtracted from max_length (happens when BOS/EOS is already present).
Fixes #2087
Before submitting
Pull Request section?
to it if that's the case.
[documentation guidelines). (not needed?)
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.