inkoziev/Conversations

这个数据集包含了大约900万条俄语对话，这些对话来源于不同的渠道，包括笑话、文学作品、在线论坛以及一小部分合成对话。每个对话都标记有通用的来源标识符，以便进行有针对性的过滤。数据集已经进行了基本预处理，包括去重、清洗、排版规范化以及小型的拼写错误修正。每个会话的转折点以斜线符号开始。

This dataset contains approximately 9 million Russian-language dialogues collected from diverse sources, including jokes, literary works, online forums, and a small portion of synthetic dialogues. Each dialogue is tagged with a generalized source identifier for targeted filtering. The dataset has undergone basic preprocessing, including deduplication, cleaning, typographical normalization, and minor typo correction. Each conversation turn begins with an em dash symbol.