mlfoundations-dev/global_batchsize_512_lradjusted8_1743496156_eval_0981
收藏Hugging Face2025-04-01 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/global_batchsize_512_lradjusted8_1743496156_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测试集的预计算模型输出，用于评估模型在数学题目上的表现。评估结果包括AIME24、AIME25、AMC23、MATH500、GPQADiamond和LiveCodeBench等多个数据集或测试集的准确率以及其他统计信息。
This dataset contains precomputed model outputs for evaluation on various test sets, assessing the models performance on mathematical questions. The evaluation results include accuracy and other statistics for multiple datasets or test sets such as AIME24, AIME25, AMC23, MATH500, GPQADiamond, and LiveCodeBench.
提供机构：
mlfoundations-dev
