mlfoundations-dev/seed_math_tiger_lab_math

该数据集包含对话信息，每个样本包括源文本(source)、响应种子(response_seed)、指令种子(instruction_seed)、响应(response)、原始行索引(__original_row_idx)以及会话信息(conversations)。会话信息由对话的发起者(from)和对话内容(value)组成。数据集被划分为训练集，共有50000个示例，文件大小为206714516字节。

The dataset contains conversation information, with each sample including source text (source), response seed (response_seed), instruction seed (instruction_seed), response (response), original row index (__original_row_idx), and conversation information (conversations). The conversation information consists of the initiator (from) and the content of the conversation (value). The dataset is split into a training set with a total of 50,000 examples, with a file size of 206,714,516 bytes.