open-r1/OpenR1-Math-220k
收藏Hugging Face2025-02-12 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/open-r1/OpenR1-Math-220k
下载链接
链接失效反馈官方服务：
资源简介：
OpenR1-Math-220k是一个包含220k个数学问题的数学推理大规模数据集。每个问题由DeepSeek R1模型生成了2到4个推理轨迹，这些轨迹大部分经过了Math Verify的验证，少部分由Llama-3.3-70B-Instruct模型进行评判。数据集分为两个部分：default和extended，其中default包含94k个问题，extended包含131k个样本，后者加入了如cn_k12等额外数据源的样本。数据集根据Apache 2.0许可证发布。
OpenR1-Math-220k is a large-scale dataset for mathematical reasoning containing 220k math problems, each with two to four reasoning traces generated by the DeepSeek R1 model. These traces have been verified by Math Verify for most samples and by Llama-3.3-70B-Instruct for 12% of the samples. The dataset is divided into two parts: default with 94k problems and extended with 131k samples, including additional data sources like cn_k12. The dataset is released under the Apache 2.0 license.
提供机构：
open-r1
