AIPlans/ultrafeedback_binarized_chinese

该数据集包含了一系列特征，如提示文本(prompt)、提示ID(prompt_id)、选择的选项(chosen)、被拒绝的选项(rejected)、消息列表(messages，包括消息内容和角色)、选择得分(score_chosen)和拒绝得分(score_rejected)。数据集分为训练集(train_prefs)和测试集(test_prefs)，其中训练集包含13583个示例，大小为54228386字节；测试集包含432个示例，大小为1645550字节。数据集的总下载大小为33483432字节，解压后的大小为55873936字节。

The dataset includes features such as prompt text (prompt), prompt ID (prompt_id), chosen option (chosen), rejected option (rejected), message list (messages including message content and role), chosen score (score_chosen), and rejected score (score_rejected). The dataset is split into a training set (train_prefs) and a test set (test_prefs), with the training set containing 13,583 examples and a size of 54,228,386 bytes; the test set contains 432 examples and a size of 1,645,550 bytes. The total download size of the dataset is 33,483,432 bytes, and the size after decompression is 55,873,936 bytes.