Bmingg/DPO_2b_vinaiv2_3600_constrained_MBR_bleurt

该数据集包含三个字段：prompt（提示）、chosen（选择的回答）和rejected（被拒绝的回答）。数据集仅包含训练集部分，共有3600个示例。数据集适用于训练选择型回答模型。The dataset includes three fields: prompt (prompt), chosen (selected response) and rejected (rejected response). The dataset contains only the training set, with a total of 3600 examples. The dataset is suitable for training selection-type response models.