mlfoundations-dev/openthoughts_113957_1742860411_eval_0981
收藏Hugging Face2025-03-25 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/openthoughts_113957_1742860411_eval_0981
下载链接
链接失效反馈官方服务：
资源简介：
这是一个包含预计算模型输出的评估数据集，用于测试模型在不同数学题目测试集上的表现。数据集包含AIME24、AIME25、AMC23、MATH500、GPQADiamond和LiveCodeBench等多个测试集，每个测试集都有多次运行的结果，提供了准确度、解题数和总题数等详细信息。
This is an evaluation dataset containing precomputed model outputs to test the performance of models on different mathematical question sets. The dataset includes multiple test sets such as AIME24, AIME25, AMC23, MATH500, GPQADiamond, and LiveCodeBench, each with multiple runs providing details on accuracy, number of questions solved, and total number of questions.
提供机构：
mlfoundations-dev
