mlfoundations-dev/Bespoke-Stratos-32B_1744004628_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/Bespoke-Stratos-32B_1744004628_eval_0981
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是mlfoundations-dev/Bespoke-Stratos-32B_1744004628_eval_0981，包含了针对六个不同任务的预计算模型输出，用于评估模型性能。这些任务包括AIME24、AIME25、AMC23、MATH500、GPQADiamond和LiveCodeBench，每个任务都有相应的平均准确率和详细评估结果。
This dataset, mlfoundations-dev/Bespoke-Stratos-32B_1744004628_eval_0981, contains precomputed model outputs for evaluation on six different tasks: AIME24, AIME25, AMC23, MATH500, GPQADiamond, and LiveCodeBench. Each task has its corresponding average accuracy and detailed evaluation results.
提供机构：
mlfoundations-dev
