The Python dataset (python_dataset.jsonl) contains 285,249 samples, while the Java dataset (java_dataset.jsonl) contains 221,795 samples. Each sample has the following structure: <index, Human-code, ChatGPT-code, DeepSeek-Coder-code, Qwen-Coder-code>.
