petkopetkov/oasst2_bg
收藏Hugging Face2024-12-04 更新2024-12-14 收录
下载链接：
https://hf-mirror.com/datasets/petkopetkov/oasst2_bg
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是OpenAssistant Conversations Dataset (OASST2)的保加利亚语翻译版本，使用了NLLB模型进行翻译。数据集包含多个特征，如消息ID、父消息ID、用户ID、创建日期、文本内容、角色、语言、评论数量、评论结果、是否删除、排名、是否为合成数据、模型名称、毒性分析、消息树ID、树状态、表情符号和标签等。数据集分为训练集和验证集，分别包含119,535和4,509个样本。
The OpenAssistant Conversations Dataset (OASST2) translated to Bulgarian language using the NLLB model. The dataset includes features such as message_id, parent_id, user_id, created_date, text, role, lang, review_count, review_result, deleted, rank, synthetic, model_name, detoxify structure (containing various toxicity-related metrics), message_tree_id, tree_state, emojis, and labels. The dataset is split into validation and train sets, containing 4509 and 119535 samples respectively.
提供机构：
petkopetkov
