Question about the evaluation dataset

huggingface / alignment-handbook

Robust recipes to align language models with human and AI preferences

https://huggingface.co/HuggingFaceH4

Apache License 2.0

4.6k stars 401 forks source link

Open chenweixin107 opened 11 months ago

chenweixin107 commented 11 months ago

Hi, I wonder which TruthfulQA task you are focusing on during evaluation? MC1, MC2, or generation task?

wj210 commented 9 months ago

asking about this as well.