zhenghaoxu/think-rm-rmr1-helpsteer3
收藏Hugging Face2025-10-21 更新2025-10-25 收录
下载链接:
https://hf-mirror.com/datasets/zhenghaoxu/think-rm-rmr1-helpsteer3
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含了文本相关的多个字段,主要用于训练和验证。具体包括:提示内容(prompt)和角色(role),数据来源(data_source),以及奖励模型(reward_model)的相关信息,如真实标签(ground_truth)和风格(style)。数据集分为训练集和验证集,其中训练集包含109265个示例,大小为710,471,346字节;验证集包含2017个示例,大小为15,206,478字节。
The dataset includes multiple text-related fields primarily used for training and validation. It consists of prompt content and role, data source, and reward model-related information such as ground truth and style. The dataset is split into a training set with 109,265 examples, totaling 710,471,346 bytes, and a validation set with 2,017 examples, totaling 15,206,478 bytes.
提供机构:
zhenghaoxu



