bench-temp/mmlu-pt-high-school-ablation-v2-notink-unbalanced
收藏数据链接：
官方服务：
资源简介：
该数据集是一个多学科问题集合，包含健康与生物科学、人文学科、语言和STEM四个领域的子集。每个数据样本包括原始任务描述、唯一问题标识符、难度正确数和难度评分。数据集总计2042个样本，旨在评估或分析不同学科领域的问题难度和性能。
This dataset is a multi-disciplinary question collection comprising subsets in health and biological sciences, humanities, languages, and STEM. Each data sample includes raw task description, unique question identifier, difficulty correct count, and difficulty score. The dataset totals 2042 samples and is designed for evaluating or analyzing question difficulty and performance across different academic domains.
提供机构：
bench-temp
