ai2-adapt-dev/tool-use-synthetic-gpt-4.1-test-fixed

该数据集包含会话信息，每个会话记录包括消息ID、消息内容、功能调用信息、功能列表、角色、消息来源、会话轮数、步骤数、执行类型、是否拒绝执行、是否可解析、API中是否有对应功能、参数数量是否正确、参数名称是否正确以及代码是否正确等字段。数据集分为训练集，其大小为279208字节，包含79个示例。

The dataset contains session information, with each session record including message ID, message content, function call information, function list, role, message source, number of turns in the session, step number, execution type, whether to refuse execution, whether it is parseable, whether there is a corresponding function in the API, whether the number of arguments is correct, whether the argument names are correct, and whether the code is correct. The dataset is split into a training set, which is 279208 bytes in size and contains 79 examples.