Direct Preference Optimization (DPO) style rewards#99
Merged
Conversation
steffencruz
reviewed
Jul 20, 2023
Contributor
steffencruz
left a comment
steffencruz
left a comment
There was a problem hiding this comment.
Interesting changes. Looking forward to seeing the experiments
Eugene-hu
approved these changes
Aug 24, 2023
Merged
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Direct Preference Optimization (DPO) style rewards
Calculates a direct preference optimization (DPO) style reward for a completion, which is a reference model's average log-probability for completion tokens given a prompt.
Uses guidance from https://github.com/eric-mitchell/direct-preference-optimization/blob/main/trainers.py.