deep-analysis-research/simple-evals
收藏Hugging Face2025-09-23 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/deep-analysis-research/simple-evals
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含多个配置，如AIME2024、AIME2025、GPQA-diamond、MATH、MATH500以及针对不同语言的多个MMLU配置。每个配置都有一个测试集分割，并提供了相应的JSONL文件路径。 README文件未提供数据集的具体描述，因此无法确定数据集的具体内容。从文件扩展名和配置名称推测，这可能是一系列用于不同任务和语言的测试数据集。
The dataset consists of multiple configurations such as AIME2024, AIME2025, GPQA-diamond, MATH, MATH500, and multiple MMLU configurations for different languages. Each configuration includes a test split and the corresponding JSONL file path. The README does not provide a specific description of the datasets, hence the exact content of the datasets is unclear. It is speculated that this might be a series of test datasets for different tasks and languages based on the file extension and configuration names.
提供机构：
deep-analysis-research
