arcee-train/arenasim-mix-unfiltered

该数据集包含对话信息，每个对话包括发送者(from)和对话内容(value)，此外还有数据来源(source)信息。数据集分为训练集，共有超过两百十五万的样本，总大小超过7GB。数据集适用于自然语言处理中的对话系统、信息提取等任务。

The dataset contains conversation information, including the sender (from) and the content of the conversation (value), as well as the source (source) information. The dataset is split into a training set with a total of over two million and one hundred fifty thousand samples, totaling more than 7GB in size. The dataset is suitable for tasks such as conversational systems and information extraction in natural language processing.