mlfoundations-dev/stackoverflow_500000_samples
收藏官方服务：
资源简介：
这是一个包含 instruction 和 completion 两个字符串字段的文本数据集，还有一个包含对话信息的 conversations 列表字段。数据集分为训练集，共有50万个样本，总大小约为3.4GB。
This is a text dataset containing two string fields, instruction and completion, as well as a list field containing conversation information. The dataset is split into a training set with a total of 500,000 samples, totaling approximately 3.4GB in size.
提供机构：
mlfoundations-dev
