mlfoundations-dev/math-stratos-verified-scaled-1

该数据集包含问题、推理过程、DeepSeek系统解决方案、真实解决方案、正确性标记、判断推理、系统信息和对话信息。对话信息包括对话来源和对话内容。数据集仅包含训练集split，共有89120个示例。

The dataset includes problem, reasoning process, DeepSeek system solutions, ground truth solutions, correctness labels, reasoning judgments, system information, and conversation information. The conversation information includes the source and content of the conversation. The dataset contains only the training set split, with a total of 89120 examples.