mothnaZl/self_rewarding_ift_Qwen2.5-7B-Instruct_prev
收藏Hugging Face2025-04-06 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/mothnaZl/self_rewarding_ift_Qwen2.5-7B-Instruct_prev
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含对话信息，每个样本包含目标标签(gt)、对话内容(conversations)、奖励值(reward)和对话长度(len)。对话内容进一步分为对话文本(content)和角色(role)。数据集提供了一个训练集分割，可用于模型训练。
The dataset includes conversation information, with each sample containing a target label (gt), conversation content (conversations), a reward value (reward), and the length of the conversation (len). The conversation content is further divided into conversation text (content) and role (role). The dataset provides a training set split for model training.
提供机构：
mothnaZl
