Recommended by our colleagues in the data science team:
📘 The LLM Evaluation guidebook
This reference (on GitHub) helps you make sure an LLM performs well on a specific task. It addresses the different ways of evaluating a model and offers guidance on evaluation design as well as practical tips.
Of course, new collaborators are welcome.
https://github.com/huggingface/evaluation-guidebook#AI #LLMs #guide
