PJMixers-Dev/grimulkan_PIPPA-augmented-dedup-system-qwq-all-aphrodite
收藏Hugging Face2025-03-12 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/PJMixers-Dev/grimulkan_PIPPA-augmented-dedup-system-qwq-all-aphrodite
下载链接
链接失效反馈官方服务：
资源简介：
该数据集被设置为类似于R1多轮对话的训练方式，其特点是在训练时应该只关注最后一个对话轮次，目的是训练模型能够总是输出思考，不受用户输入内容的影响。
The dataset is set up to be trained like R1 multi-round conversation, where only the last turn of the conversation should be focused on during training, aiming to teach the model to always output thinking regardless of the users input.
提供机构：
PJMixers-Dev
