ajibawa-2023/Cpp-Code-Large

Cpp-Code-Large Cpp-Code-Large is a large-scale corpus of C++ source code comprising more than 5 million lines of C++ code. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and static program analysis for the C++ ecosystem. By providing a high-volume, language-specific corpus, Cpp-Code-Large enables systematic experimentation in C++-focused model training, domain adaptation, and downstream code… See the full description on the dataset page: https://huggingface.co/datasets/ajibawa-2023/Cpp-Code-Large.