mlfoundations-dev/Qwen2.5-7B-Instruct_qwq_mix_r1_science_eval_2870
收藏Hugging Face2025-06-30 更新2025-07-05 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/Qwen2.5-7B-Instruct_qwq_mix_r1_science_eval_2870
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AIME24评估的预计算模型输出结果。评估结果显示了10次运行的准确率，平均准确率为60.67%，每次运行回答的问题数量在14到21之间，总问题数量为30。
This dataset contains precomputed model outputs for the AIME24 evaluation. The evaluation results show accuracy for 10 runs with an average accuracy of 60.67%, and the number of questions answered ranges from 14 to 21 out of 30 total questions.
提供机构：
mlfoundations-dev
