nbalepur/open-llm-benchmark
收藏Hugging Face2025-04-04 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/nbalepur/open-llm-benchmark
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含四个子数据集：ARC、HellaSwag、gsm8k和mmlu。每个子数据集都包括问题和对应的选项，以及正确的答案字母。ARC和HellaSwag子数据集有训练集和测试集，gsm8k和mmlu子数据集同样包含训练集和测试集，但是gsm8k的选项信息为null。数据集的具体大小和下载大小也已经给出。
The dataset consists of four sub-datasets: ARC, HellaSwag, gsm8k, and mmlu. Each sub-dataset includes questions with corresponding choices and the correct answer letter. Both ARC and HellaSwag sub-datasets have training and test sets, while gsm8k and mmlu also contain training and test sets but the choices information for gsm8k is null. The specific size of the dataset and the download size are provided.
提供机构：
nbalepur
