TAUR-dev/evals__full_TAUR_dev__convos__long_multiplication__four_digit__train__r1__samples

该数据集包含了文档ID、文档内容（包括答案、问题、问题和解决方案）、目标、参数（包括生成参数和采样参数等）、响应和过滤后的响应、文档哈希、提示哈希、目标哈希、精确匹配标志和提取的答案等字段。数据集分为训练集，共有100个示例，总大小为3,304,076字节。数据集的下载大小为1,084,486字节。

The dataset includes fields such as document ID, document content (including answer, problem, question, and solution), target, arguments (including generation arguments and sampling parameters), responses and filtered responses, document hash, prompt hash, target hash, exact match flag, and extracted answers. The dataset is split into a training set with a total of 100 examples and a size of 3,304,076 bytes. The download size of the dataset is 1,084,486 bytes.