HuggingFaceH4/prm800k-trl-dedup
收藏Hugging Face2025-01-09 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/HuggingFaceH4/prm800k-trl-dedup
下载链接
链接失效反馈官方服务：
资源简介：
这是一个用于微调PRM模型的数据集，包含了问题情景、针对情景的不同解答步骤以及每个步骤的正确性标签。数据集基于tasksource/PRM800K创建，并经过去重处理，分为训练集和测试集。
This dataset is prepared for fine-tuning PRM models, containing problem scenarios, different solution steps for each scenario, and labels indicating the correctness of each step. It is based on the tasksource/PRM800K dataset and has been deduplicated, split into training and test sets.
提供机构：
HuggingFaceH4
