Known

该数据集旨在分析语言模型中特征归因方法的忠实度，其中每个实例包含了例如“勒布朗·詹姆斯职业从事的运动是[篮球]”这样的信息。该数据集包含了1000个这样的实例，其任务是进行特征归因分析。

This dataset is designed to analyze the faithfulness of feature attribution methods in language models. Each instance contains information such as "LeBron James' professional sport is [basketball]". The dataset consists of 1000 such instances, with the task focused on feature attribution analysis.