YuchenLi01/MATH_1.5Bsft_Score_DPO_Qwen2.5MathRM72B_hard0soft16_all_soft_random_unfiltered

该数据集包含四个字段：prompt（提示语）、chosen（选择的答案）、rejected（被拒绝的答案）和all_rm_scores（所有拒绝答案的得分）。数据集分为训练集和测试集，训练集包含113584个示例，测试集包含6416个示例。数据集总大小约为1.09GB。The dataset includes four fields: prompt (prompt text), chosen (selected answer), rejected (rejected answer), and all_rm_scores (scores of all rejected answers). The dataset is split into a training set and a test set, with the training set containing 113584 examples and the test set containing 6416 examples. The total size of the dataset is approximately 1.09GB.