TAUR-dev/agg__BC__math500_v3_distilled

该数据集包含四个浮点型特征：回溯(backtracking)、向后链接(backward_chaining)、验证(verification)和子目标(subgoals)。数据集仅包含一个训练集分割，包含1个样本，总共32字节。数据集的下载大小为2067字节。

The dataset includes four floating-point features: backtracking, backward_chaining, verification, and subgoals. The dataset contains only one training split with 1 example, totaling 32 bytes. The download size of the dataset is 2067 bytes.