Skip to content
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.

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

Support multiple completions for ModelbasedClassify #1484

Open
tom-christie opened this issue Mar 14, 2024 · 0 comments
Open

Support multiple completions for ModelbasedClassify #1484

tom-christie opened this issue Mar 14, 2024 · 0 comments

Comments

@tom-christie
Copy link

tom-christie commented Mar 14, 2024

Describe the feature or improvement you're requesting

It would be nice to be able to score multiple sample completions using ModelBasedClassify. Even if n>1 is passed into a completion function and multiple samples are returned, only the first is graded because of this line:

https://github.com/openai/evals/blob/main/evals/elsuite/utils.py#L193

Additional context

I would like to be able to raise the temperature, ask a model to produce N completions, and have each completion graded separately using a rubric. This appears to work fine for non-model-based scoring.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant