mlfoundations-dev/AIME24_evalchemy_16shards_Qwen2.5-7B-Instruct
收藏Hugging Face2025-03-07 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mlfoundations-dev/AIME24_evalchemy_16shards_Qwen2.5-7B-Instruct
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AIME24评估。数据集包含了5次运行的评估结果，每次运行有30个问题，记录了每次运行的准确率、解决的问题数和总问题数。平均准确率为10.67%，标准差为2.56%。
This is an evaluation dataset containing precomputed model outputs for AIME24 assessment. The dataset includes the evaluation results of 5 runs, each with 30 questions, documenting the accuracy, number of questions solved, and total number of questions for each run. The average accuracy is 10.67% with a standard deviation of 2.56%.
提供机构：
mlfoundations-dev
