artianand/bbq_roberta_base_race_custom_loss_predictions

该数据集是一个包含多个字段的结构化数据集，主要用于训练机器学习模型。它包含了类别、示例ID、问题索引、问题极性、上下文条件、上下文、问题、三个答案选项、答案信息、答案标签、目标标签以及附加元数据等信息。数据集分为训练集，共有54992个示例，总大小为44065722字节。附加元数据中包含了关于答案对齐、条件、已知刻板印象群体、标签类型、相关社会价值观、来源、刻板印象群体、子类别和版本等信息。

The dataset is a structured dataset with multiple fields designed for training machine learning models. It includes information such as category, example ID, question index, question polarity, context condition, context, question, three answer options, answer information, answer label, target label, and additional metadata. The dataset is split into a training set with a total of 54,992 examples and a size of 44,065,722 bytes. The additional metadata includes details about answer alignment, conditions, known stereotyped groups, label type, relevant social values, source, stereotyped groups, subcategory, and version.