Dan Saattrup Smart
Dan Saattrup Smart
World Values Survey: https://www.worldvaluessurvey.org/WVSContents.jsp
This dataset might be relevant: https://openai.com/index/introducing-simpleqa/
Could utilise this language detection tool, which seems to be the most accurate: https://github.com/pemistahl/lingua-py
This is a bug in outlines, which is used when benchmarking NER tasks. I'll reach out to them, but you can benchmark all other tasks at least!
Opened PR in Outlines to fix this now: https://github.com/dottxt-ai/outlines-core/pull/90
> Opened PR in Outlines to fix this now: [dottxt-ai/outlines-core#90](https://github.com/dottxt-ai/outlines-core/pull/90) Update: This PR has now been merged in, so this should be fixed when we get a new `outlines-core` release,...
Another update! It has now been released in `outlines-core`: https://github.com/dottxt-ai/outlines-core/releases/tag/0.1.18 However, `outlines` currently doesn't allow that version yet, so waiting for a new release that fixes that...
Outlines has updated its `outlines-core` dependency now. The last step is that vLLM should allow a newer version of `outlines`, which will happen when this PR is merged in and...
> Outlines has updated its `outlines-core` dependency now. The last step is that vLLM should allow a newer version of `outlines`, which will happen when this PR is merged in...
Thanks for your feedback! Is it correctly understood that you're asking for both the primary metric score and the secondary metric score to be included in the CSV export?