Nutanix/kto-mix-test

该数据集包含四个主要特征：prompt、completion、label和__index_level_0__。其中，prompt和completion都是列表类型，包含content和role两个字段，类型均为字符串。label字段为布尔类型，__index_level_0__字段为整型。数据集分为train和test两个部分，每个部分的大小和样本数量相同。数据集的下载大小和总大小也在README中给出。

The dataset contains four main features: prompt, completion, label, and __index_level_0__. Among them, prompt and completion are both list types, containing two fields: content and role, both of which are string types. The label field is of boolean type, and the __index_level_0__ field is of integer type. The dataset is divided into two parts: train and test, with each part having the same size and number of samples. The download size and total size of the dataset are also provided in the README.