bethrezen/gpt-ru-reasoning_effort-sft

该数据集包含对话信息，每个对话示例由prompt_id标识，包含发送者的角色(role)、消息内容(content)和思考过程(thinking)。此外，每个示例还标注了语言(language)和语言匹配度(language_match)。数据集分为训练集，共有约159558个示例，数据大小为约1GB。

The dataset consists of conversational information, with each conversation example identified by a prompt_id, including the senders role (role), message content (content), and thought process (thinking). Additionally, each example is annotated with the language (language) and language match degree (language_match). The dataset is split into a training set, containing approximately 159,558 examples, with a total data size of about 1GB.