knoveleng/redbench

该数据集包含多个配置，每个配置都包含了一系列的特征，如提示语(prompt)、选项(choices)、答案(answer)等，这些都是字符串类型的数据。此外，还包括任务类型(task)、子任务(subtask)、分类(category)、领域(domain)、语言(language)和来源(source)等特征，用于详细描述数据集的属性。每个数据集还有一个训练集(train)的分割，提供了该训练集的大小和包含的示例数量。

This dataset comprises multiple configurations, each with a set of features such as prompt, choices, answer, all of which are string-typed data. Additionally, it includes features like task type, subtask, category, domain, language, and source, providing detailed descriptions of the datasets attributes. Each dataset also has a train split, detailing the size of the training set and the number of examples it contains.