💔 [SFT] Raise error when formatting_func is used with completion_only_loss
#3385
Conversation
This file contains hidden or 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
qgallouedec
reviewed
Apr 29, 2025
qgallouedec
reviewed
Apr 29, 2025
qgallouedec
reviewed
Apr 29, 2025
trl/trainer/sft_trainer.py
Outdated
Comment on lines
344
to
345
| "which is incompatible with completion-only loss. The completion_only_loss is True either because " | ||
| "you explicitly set it or because your dataset is in prompt-completion format." |
There was a problem hiding this comment.
Can you incorporate the following recommendation? I think it's more actionable:
apply your format function before passing your dataset, or alternatively, disable
completion_only_lossinSFTConfig.
There was a problem hiding this comment.
Have a look now and tell me what you think
formatting_func + completion_only_lossformatting_func is used with completion_only_loss
formatting_func is used with completion_only_lossformatting_func is used with completion_only_loss
|
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Outdated suggestions cannot be applied.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?
Succeeds #3381. Instead of extending
formatting_functo support prompt-completion we update the documentation and raise an error whenever a formatting function is passed together withcompletion_only_loss=TrueFixes #3343
Before submitting
Pull Request section?
to it if that's the case.
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.