mlfoundations-dev/OpenR1-Math-Raw-all-correct-sharegpt_1744357980_eval_7e69
收藏Hugging Face2025-04-11 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/OpenR1-Math-Raw-all-correct-sharegpt_1744357980_eval_7e69
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了数学问题的预计算模型输出，用于评估模型在FinanceBench测试上的表现，其中平均准确度为70.00%，仅运行了一次评估。
This dataset includes precomputed model outputs for math questions, used for evaluating the models performance on the FinanceBench test, with an average accuracy of 70.00%, and only one evaluation run was conducted.
提供机构：
mlfoundations-dev
