mlfoundations-dev/seed_math_nvidia_math

该数据集包含了七个字段：指令种子(instruction_seed)、生成解决方案(generated_solution)、响应种子(response_seed)、问题来源(problem_source)、响应(response)、原始行索引(__original_row_idx)和对话(conversations)。对话字段是一个列表，记录了对话的来源和内容。数据集分为训练集，共有50000个示例，大小为231,389,928字节。数据集的总下载大小为108,367,277字节。

The dataset includes seven fields: instruction_seed, generated_solution, response_seed, problem_source, response, __original_row_idx, and conversations. The conversations field is a list that records the source and content of the conversation. The dataset is split into a training set with a total of 50,000 examples, totaling 231,389,928 bytes in size. The total download size of the dataset is 108,367,277 bytes.