mlfoundations-dev/oh_teknium_scaling_down_ratiocontrolled_0.6

该数据集包含对话信息和一个标签序列。对话信息由消息来源(from)、消息内容(value)和权重(weight)组成。数据集划分为训练集，共有600,930个示例，大小为约976.64MB。数据集还提供了默认配置信息，指定了训练集的数据文件路径。

The dataset consists of conversation information and a label sequence. The conversation information is composed of message source (from), message content (value), and weight (weight). The dataset is split into a training set with a total of 600,930 examples, totaling approximately 976.64MB. The dataset also provides default configuration information, specifying the data file path for the training set.