ioi-leaderboard/ioi-eval-sglang_Qwen_Qwen2.5-1.5B-Instruct-new

该数据集包含了编程问题及其相关信息的记录，每个问题都有唯一的标识符（problem_id）、子任务类型（subtask）、问题提示（prompt）、生成文本（generation）、代码（code）、编程语言（language）、解决方案数量（solution_number）以及一个唯一标识符（uuid）。此外，还包含了模型参数（model_kwargs）和元数据（metadata），其中元数据提供了关于使用情况（如令牌数量和成本）和时间戳的信息。数据集分为训练集，共有2050个示例，大小为22012353字节。

The dataset consists of records for programming problems and their associated information, each with a unique identifier (problem_id), subtask type (subtask), prompt for the problem (prompt), generated text (generation), code (code), programming language (language), number of solutions (solution_number), and a unique identifier (uuid). It also includes model arguments (model_kwargs) and metadata, which provide information about usage (such as token counts and cost) and a timestamp. The dataset is split into a training set with a total of 2050 examples, with a size of 22012353 bytes.