mlfoundations-dev/stackoverflow_375000_samples

该数据集包含四个字段：指令（instruction）、完成情况（completion）、原始行索引（__original_row_idx）和会话信息（conversations）。会话信息中又包括消息来源（from）和消息内容（value）。数据集仅包含一个训练集，共有375000个示例。数据集的总大小为2,682,969,752字节。The dataset includes four fields: instruction, completion, __original_row_idx, and conversations. The conversations field contains information about the source (from) and content (value) of messages. The dataset consists of only one training set with a total of 375,000 examples. The total size of the dataset is 2,682,969,752 bytes.