mlfoundations-dev/SCP_40k_R1_with_OT_unverified_eval_03-11-25_06-25-08_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_unverified_eval_03-11-25_06-25-08_f912
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是SCP_40k_R1_with_OT_unverified_eval_03-11-25_06-25-08_f912的预计算模型输出评估数据，包含3次运行的评估结果，平均准确度为27.61%。每次运行评估了198个问题，涉及准确度、解决的问题数量等信息。
This dataset is the precomputed model outputs for evaluation of SCP_40k_R1_with_OT_unverified_eval_03-11-25_06-25-08_f912, containing the evaluation results of 3 runs with an average accuracy of 27.61%. Each run evaluated 198 questions, involving accuracy, number of questions solved, and other information.
提供机构：
mlfoundations-dev
