marcov/glue_mnli_promptsource

该数据集包含多个特征，如前提（premise）、假设（hypothesis）、标签（label）等，标签有三种类别：entailment、neutral和contradiction。数据集分为训练集、验证集（匹配和不匹配）和测试集（匹配和不匹配）。每个分割的样本数量和字节大小也被提供。数据集的总下载大小和总大小也被列出。

The dataset contains multiple features such as premise, hypothesis, label, etc., with labels categorized into entailment, neutral, and contradiction. The dataset is divided into training, validation (matched and mismatched), and test sets (matched and mismatched). The number of samples and byte size for each split is also provided. The total download size and total size of the dataset are also listed.