sartifyllc/orca-agentinstruct-1M-v1

该数据集包含多个分割，每个分割都有特定的主题和任务，如创意内容、文本修改、结构到文本流程、阅读理解、检索增强生成、文本提取、多项选择题、后续问题、分析推理、费米问题、流程推理、代码、脑筋急转弯、文本分类和开放域问答。每个分割都包含大量的对话数据，对话数据由messages和conversations组成，其中conversations包含对话的来源和内容。数据集的总下载大小为4384035156字节，总数据集大小为9441987720字节。

This dataset contains multiple splits, each with specific themes and tasks such as creative content, text modification, structure-to-text flow, reading comprehension, retrieval-augmented generation, text extraction, multiple-choice questions, follow-up questions, analytical reasoning, Fermi problems, flow reasoning, code, brain teasers, text classification, and open-domain question answering. Each split contains a large amount of conversational data, which consists of messages and conversations, where conversations include the source and content of the dialogue. The total download size of the dataset is 4384035156 bytes, and the total dataset size is 9441987720 bytes.