mlfoundations-dev/SYNTHETIC-1_verified_100k_batch_size_512_1743212172_eval_0771
收藏Hugging Face2025-03-29 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/SYNTHETIC-1_verified_100k_batch_size_512_1743212172_eval_0771
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是预计算出的模型输出，用于评估数学问题解答模型的性能。它包含了三个不同的评估指标：AIME24，AIME25，和MATH500，每个指标都有多次运行的平均准确度结果。
This dataset consists of precomputed model outputs for the purpose of evaluating the performance of mathematical problem-solving models. It includes three different evaluation metrics: AIME24, AIME25, and MATH500, each with average accuracy results from multiple runs.
提供机构：
mlfoundations-dev
