mlfoundations-dev/OpenThinker-7B_1743586838_eval_f912
收藏Hugging Face2025-04-02 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/OpenThinker-7B_1743586838_eval_f912
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了针对GPQADiamond任务预计算出的模型输出，用于评估。数据集中提供了平均准确率、运行次数以及每次运行的准确率、解决的问题数量和总问题数量。
This dataset contains precomputed model outputs for evaluation, specifically for the GPQADiamond task. It includes the average accuracy, number of runs, and detailed results for each run such as accuracy, number of questions solved, and total questions.
提供机构：
mlfoundations-dev
