mlfoundations-dev/sky_t1_data_8k

该数据集包含了对话系统中的交互信息。具体来说，每个样本包含了一个system特征和一个conversations列表特征，其中system表示系统的话语，conversations列表中的from子特征表示说话者，value子特征表示说话内容。数据集分为训练集(train)，共有8000个示例，大小为123575833字节。

The dataset contains interaction information in dialogue systems. Specifically, each sample includes a system feature and a conversations list feature, where the system feature represents the systems utterance, and the conversations lists from subfeature represents the speaker, while the value subfeature represents the content of the speech. The dataset is split into a training set (train) with a total of 8000 examples, totaling 123575833 bytes in size.