mlfoundations-dev/star_coder_instruct

该数据集包含对话信息，每个示例包括响应种子(response_seed)、指令种子(instruction_seed)、响应(response)以及原始行索引(__original_row_idx)。此外，每个示例还包含一个对话列表(conversations)，其中包含对话的发起者(from)和对话内容(value)。训练集包含50000个示例，数据集大小为269,654,396字节。

The dataset includes conversation information, with each example containing a response seed (response_seed), instruction seed (instruction_seed), response (response), and original row index (__original_row_idx). Additionally, each example includes a conversation list (conversations) that contains the initiator (from) and content (value) of the conversation. The training set contains 50,000 examples, and the dataset size is 269,654,396 bytes.