VGraf/mt_dependent_assistant_2_turns

该数据集是一个对话数据集，包含用户(user)和系统(sys)之间的对话，以及对话的标识(id)，是否执行推理(do_inference)，指令(inst)，关键字(key)和提示(prompt)。数据集分为训练集(train)，包含600个示例。此外，数据集还提供了两种配置，用于指定不同的测试和训练数据文件。

This dataset is a dialogue dataset containing conversations between users (user) and systems (sys), along with identifiers (id), whether to perform inference (do_inference), commands (inst), keywords (key), and prompts (prompt). The dataset is split into a training set (train) with 600 examples. Additionally, the dataset provides two configurations for specifying different test and training data files.