bench-temp/mmlu-pt-undergraduate-difficulty-v2-notink
收藏官方服务：
资源简介：
该数据集包含多个学科（会计、建筑、工商管理、经济学、健康与生物科学、人文学科、语言、法律、科学、技术、工程和数学）的问题。每个问题包含原始任务文本、问题ID、难度相关指标（包括正确次数、错误次数、难度分数、难度分类）以及正确和错误模型列表。
This dataset contains questions from multiple disciplines (accounting, architecture, business administration, economics, health and biological sciences, humanities, languages, law, STEM). Each question includes raw task text, question ID, difficulty metrics (number of correct answers, number of wrong answers, difficulty score, difficulty bin), and lists of models that answered correctly or incorrectly.
提供机构：
bench-temp
