dsrtrain/raft_iter2_sft
收藏Hugging Face2025-02-09 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/dsrtrain/raft_iter2_sft
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，每个样本包括对话内容(data)、地面真实标签和风格(gt)，对话中的每一次发言(conversations)以及发言的轮数(turn)。训练集(train)包含22435个样本，总数据大小为77429200字节。
The dataset consists of conversational information, with each sample including dialogue content (data), ground truth labels and style (gt), each speech in the conversation (conversations), and the turn number of the speech (turn). The training set (train) contains 22435 samples, with a total dataset size of 77429200 bytes.
提供机构：
dsrtrain
