[evals] Create some multi-turn evals
Open
area/evals
enhancement
- Dominant language
- Go
- Stars
- 569
- Forks
- 364
- Avg merge
- 2d 19h
- Merged PRs (30d)
- 136
Description
Vally supports multi-turn evals!: https://github.com/microsoft/vally/pull/495
The restrictions are that the prompts are just static strings, but other than that, we should be able to try to make some more conversation-like evals.
@tg-msft, I filed this issue as a placeholder - please edit and add in cases you think we could use this for.
Contributor guide
Assessment
This issue has not been assessed yet.