New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Unification of preference dataset format #51

Open

odelalleau opened this issue Dec 8, 2023 · 0 comments

Collaborator

odelalleau commented Dec 8, 2023

Is your feature request related to a problem? Please describe.

DPO and RM training use different dataset formats => these should be unified.

Describe the solution you'd like

Ideally we should also try to make them closer to the chat datasets, which would bring several benefits:

More consistency => simpler for users and makes it easier to write data processing scripts
Getting rid of the special formatting with extra_id's => can re-use the same dataset with different formats (e.g. extra_id vs bytes)

Details still TBD

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment