mlfoundations-dev/s1-32B_1744017028_eval_0981
收藏Hugging Face2025-04-07 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/s1-32B_1744017028_eval_0981
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了不同数学竞赛和编程竞赛的预计算模型输出结果，用于评估模型在各个任务上的表现。评估结果涵盖了AIME24、AIME25、AMC23、MATH500、GPQADiamond和LiveCodeBench等多个子数据集，每个子数据集都提供了平均准确率、每次运行的准确率、解决的问题数和总问题数。
This dataset includes precomputed model outputs for various mathematical and programming contests, used for evaluating the performance of the model on different tasks. The evaluation results cover multiple sub-datasets such as AIME24, AIME25, AMC23, MATH500, GPQADiamond, and LiveCodeBench, each providing average accuracy, accuracy per run, number of questions solved, and total number of questions.
提供机构：
mlfoundations-dev
