NJUDeepEngine/bigbench_jsonl

bigbench是一个包含众包、专家和机器生成的多语言数据集，适用于多项选择题、问答、文本分类、文本生成和零样本分类等多种任务。数据集限制在50k示例以内，以保持轻量，并且当训练集可用时，省略了默认的数据集分割。数据集遵循Apache-2.0许可证。

bigbench is a multilingual dataset with annotations and languages crowdsourced, expert-generated, and machine-generated, suitable for tasks such as multiple-choice questions, question-answering, text classification, text generation, and zero-shot classification. The dataset is limited to 50k examples to keep it lightweight, and the default split is omitted when the training set is available. The dataset is licensed under Apache-2.0.