SafeEdit encompasses 4,050 training, 2,700 validation, and 1,350 test instances.
SafeEdit can be utilized across a range of methods, from supervised fine-tuning to reinforcement learning that demands preference data for more secure responses, as well as knowledge editing methods that require a diversity of evaluation texts.… See the full description on the dataset page:
https://huggingface.co/datasets/zjunlp/SafeEdit.
