[DPO] add reference log-prob outputs in DPO#521
Merged
Tcc0403 merged 6 commits intolinkedin:mainfrom Jan 30, 2025
Merged
Conversation
3 tasks
austin362667
added a commit
that referenced
this pull request
Feb 9, 2025
## Summary <!--- This is a required section; please describe the main purpose of this proposed code change. ---> #521 introduced a unit test failure caused by the numerical issue in bf16. This PR relaxes the atol for `chosen_rewards` and `rejected_rewards` only. Note: It's just a temporary fix for ci. This PR also removes duplicated calculations on both terms. <!--- ## Details This is an optional section; is there anything specific that reviewers should be aware of? ---> ## Testing Done <!--- This is a required section; please describe how this change was tested. ---> <!-- Replace BLANK with your device type. For example, A100-80G-PCIe Complete the following tasks before sending your PR, and replace `[ ]` with `[x]` to indicate you have done them. --> - Hardware Type: <BLANK> - [ ] run `make test` to ensure correctness - [ ] run `make checkstyle` to ensure code style - [ ] run `make test-convergence` to ensure convergence --------- Co-authored-by: Austin Liu <austin362667@gmail.com>
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
4 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Since the DPO uses a reference model we also need to return the reference logprobs in DPO
Testing Done
make testto ensure correctnessmake checkstyleto ensure code stylemake test-convergenceto ensure convergence