Self-Instruct Synthetic Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/yizhongw/self-instruct
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个综合性的数据集，包含了超过52,000条指令和82,000个实例，这些数据是利用语言模型生成的，旨在提升模型遵循指令的能力。此外，该数据集使用了GPT-3的“davinci”引擎进行生成，包含了多样化的意图和文本格式。在规模上，该数据集拥有52,000条指令和82,000个实例，其任务是进行指令调优。
This is a comprehensive dataset containing over 52,000 instructions and 82,000 instances. All data is generated using language models, aiming to enhance the instruction-following capabilities of models. Additionally, this dataset was generated with the "davinci" engine of GPT-3, and covers diverse intents and text formats. In terms of scale, this dataset includes 52,000 instructions and 82,000 instances, and its target task is instruction tuning.
提供机构：
OpenAI
