code-planning/deepseek-family

该数据集包含模型在基准测试上的表现数据，具体特征包括模型名称、基准测试名称、基准测试问题ID、输入、代码输出、基础正确性、增强正确性、变体和正确性等。数据集仅包含训练集，共有7938个样本，总大小为19198447字节，下载大小为7608631字节。This dataset contains performance data of models on benchmark tests, with specific features including model name, benchmark name, benchmark question ID, input, code output, base correctness, enhanced correctness, variant, and correctness. The dataset only includes a training set, with a total of 7938 samples, a total size of 19198447 bytes, and a download size of 7608631 bytes.