limhyeonseok/NON-SKELETON-Qwen3-4B-SELF

该数据集包含三个主要字段：id、prompt和completion。其中，id字段是整数类型，用于唯一标识每个样本。prompt字段包含两个子字段：content和role，都是字符串类型，分别表示提示的内容和角色。completion字段包含三个子字段：content、reasoning_content和role，都是字符串类型，分别表示完成的文本内容、推理内容和角色。数据集分为训练集(train)，共有926个样本，大小为41796293字节。数据集的总下载大小为17216490字节。

The dataset includes three main fields: id, prompt, and completion. The id field is of integer type, used to uniquely identify each sample. The prompt field contains two sub-fields: content and role, both of which are string types, representing the content of the prompt and the role, respectively. The completion field contains three sub-fields: content, reasoning_content, and role, all of which are string types, representing the completed text content, reasoning content, and role, respectively. The dataset is split into a training set (train) with a total of 926 samples and a size of 41796293 bytes. The total download size of the dataset is 17216490 bytes.