ChavyvAkvar/aya_collection_language_split-banjar-Converted
收藏Hugging Face2025-09-05 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/ChavyvAkvar/aya_collection_language_split-banjar-Converted
下载链接
链接失效反馈官方服务：
资源简介：
这个数据集包含了对话信息，每个对话信息包括内容(content)和角色(role)两个部分，同时还记录了每个对话的token数量(num_tokens)和JSON格式的元数据(json_metadata)。数据集被划分为训练集(train)，共有3,573,365个示例，总大小为2,070,114,942字节。
The dataset consists of conversation information, with each conversation including two parts: content and role, as well as the number of tokens (num_tokens) and JSON-formatted metadata (json_metadata) for each conversation. The dataset is split into a training set (train) with a total of 3,573,365 examples and a size of 2,070,114,942 bytes.
提供机构：
ChavyvAkvar
