intelsense/openhermes-en2bn-messages-2

这是一个包含对话数据的训练集，主要用于训练对话系统。数据集中的每个样本包含源信息(source)、系统消息(system_message)、人类消息(human_message)和GPT生成的消息(gpt_message)，以及它们的bn版本，可能表示不同语言版本。数据集仅包含训练集split，共有19290个样本。

This is a training dataset containing dialogue data, primarily used for training conversational systems. Each sample in the dataset includes source information (source), system messages (system_message), human messages (human_message), and messages generated by GPT (gpt_message), along with their bn versions, which may represent different language versions. The dataset contains only the training set split, with a total of 19,290 samples.