huggingface / lighteval

Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends
MIT License
690 stars 78 forks source link

Just adding the custom metrics system #65

Closed clefourrier closed 7 months ago

clefourrier commented 7 months ago

Let's include this for the release in case I don't debug IFEval soon enough

clefourrier commented 7 months ago

Added an example :)