anakin87/tulu-3-sft-mixture-with-language
收藏Hugging Face2024-12-11 更新2024-12-14 收录
下载链接：
https://hf-mirror.com/datasets/anakin87/tulu-3-sft-mixture-with-language
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个多语言数据集，包含多种语言的文本数据，数据集的规模在10万到100万之间。数据集的来源包括多个公开的数据集，如allenai/coconot、ai2-adapt-dev/flan_v2_converted等。数据集的特征包括id、messages、source和language。这是一个带有语言列标识的tulu-3-sft-mixture数据集的版本，语言检测使用了fastText工具，但可能存在错误。
This dataset is a multilingual dataset containing text data in various languages, with a size ranging between 100K and 1M. The dataset sources include multiple publicly available datasets such as allenai/coconot, ai2-adapt-dev/flan_v2_converted, etc. The features of the dataset include id, messages, source, and language. It is a version of the tulu-3-sft-mixture dataset with a column indicating language, and language detection was performed using the fastText tool, but it may contain errors.
提供机构：
anakin87
