Dataset Card for NorTruthfulQA Generation
Dataset Details
Dataset Description
NorTruthfulQA Generation is an open-ended question answering (QA) dataset designed to evaluate whether a language model generate answers that convey false beliefs or misconceptions. The dataset contains 471 examples in both written standards of Norwegian: Bokmål and Nynorsk (the minority variant). Each example includes a question, 1 to 12 correct and incorrect reference answers, and a… See the full description on the dataset page: https://huggingface.co/datasets/ltg/nortruthfulqa_gen.
