TAUR-dev/evals__Kludge_gpt4omini_annotator

该数据集包含了文档ID、文档内容（包括答案、难度等级、问题、解决方案、学科和唯一标识符）、目标、参数（包括生成参数和采样参数）、响应和过滤后的响应、文档哈希值、提示哈希值、目标哈希值、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500个示例，总大小为5,399,418字节。

The dataset includes fields such as document ID, document content (including answer, difficulty level, problem, solution, subject, and unique identifier), target, arguments (including generation parameters 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 500 examples and a size of 5,399,418 bytes.