nuprl/MultiPL-E-completions

该数据集包含多个配置，每个配置都有特定的特征、划分和大小。数据集似乎与编程任务相关，特征包括“问题”、“语言”、“提示”、“测试”、“补全”、“程序”、“标准输出”、“标准错误”、“退出代码”、“状态”和“时间戳”。这些特征表明数据集用于评估或生成不同编程语言的代码。数据集被划分为“测试”集，每个配置具有不同数量的示例和大小。

The dataset contains multiple configurations, each with specific features, splits, and sizes. The datasets appear to be related to programming tasks, as indicated by features such as problem, language, prompt, tests, completions, programs, stdouts, stderrs, exit_codes, statuses, and timestamps. These features suggest that the datasets are used for evaluating or generating code in different programming languages. The datasets are split into test sets, with each configuration having a different number of examples and sizes.