Ayush-Singh/RM-Bench-math-gpt-4o-mini-scores-set1

该数据集包含多个字段，如唯一标识符(id)、提示(prompt)、子集(subset)等。每个样本都有三个被选中的答案(chosen_1, chosen_2, chosen_3)和三个被拒绝的答案(rejected_1, rejected_2, rejected_3)，以及相应的评分和评分理由。数据集分为训练集(train)，共有529个样本。The dataset includes multiple fields such as unique identifier (id), prompt, subset, etc. Each sample has three chosen answers (chosen_1, chosen_2, chosen_3) and three rejected answers (rejected_1, rejected_2, rejected_3), along with corresponding scores and reasons for the scores. The dataset is divided into a training set (train) with a total of 529 samples.