gswamy/pythia-1.4B-tldr-sft_tldr_pythia_1.4b_rm_sft_tldr_pythia_1.4b_quarter_2_iter_1

该数据集包含了多个序列化的整数特征和浮点数特征，具体包括最佳和最差查询响应及其掩码和奖励。数据集仅包含训练集部分，共有209580个示例，总大小为3.8GB。

The dataset includes multiple serialized integer features and floating-point features, specifically including the best and worst query responses, their masks, and rewards. The dataset contains only the training set, with a total of 209,580 examples and a total size of 3.8GB.