PJMixers/tasksource_oasst2_pairwise_rlhf_reward-PreferenceShareGPT
收藏Hugging Face2024-05-30 更新2024-06-15 收录
下载链接：
https://hf-mirror.com/datasets/PJMixers/tasksource_oasst2_pairwise_rlhf_reward-PreferenceShareGPT
下载链接
链接失效反馈官方服务：
资源简介：
---
tags:
- preference
- preferences
size_categories:
- 10K<n<100K
task_categories:
- reinforcement-learning
---
This dataset pertains to user preferences or choices, containing between 10,000 and 100,000 samples, and is suitable for reinforcement learning tasks.
提供机构：
PJMixers
原始信息汇总
数据集标签
- 偏好
- 偏好
数据集大小分类
- 10K<n<100K
任务分类
- 强化学习
