mlfoundations-dev/SCP_40k_R1_with_OT_verified_eval_03-11-25_06-25-06_f912
收藏Hugging Face2025-03-11 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/SCP_40k_R1_with_OT_verified_eval_03-11-25_06-25-06_f912
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了用于评估的预计算模型输出结果，具体评估指标为GPQADiamond，其平均准确率为30.47%。数据集记录了三次运行的结果，准确率分别为23.74%、22.73%和44.95%。
This dataset includes precomputed model outputs for evaluation purposes, with the GPQADiamond metric showing an average accuracy of 30.47%. The dataset records the results of three runs with accuracy rates of 23.74%, 22.73%, and 44.95% respectively.
提供机构：
mlfoundations-dev
