Design and Implementation of Cpp Dictionary Based on ChatGLM2-6B

Large language models play an important role in the field of education and programming assistance. In this paper, we retrained an open source LLM(ChatGLM2-6B) with Cpp programming documents, and build a web application for users to achieve specific functions of the retrained model which can offering personalized Cpp programming suggest. The system focuses on collecting Cpp-related open-source data, including API documentation, tutorials, etc., to build high-quality datasets, which are used to retrain LLM, aiming to provide accurate programming suggestions and solutions. We pre-process about 100G data, and choose high -quality question and answer in Cpp programming, finally formed 150,000 conversation records for training. And a Tauri framework based web application system is build for test and use. Experiments show that the system significantly improves Cpp programming's efficiency and quality. It can answer programming questions, optimize code, and create a fulfilling learning environment for novices. These results not only demonstrate the potential of AI in software development, but also guide future research in a wider range of languages and environments.