Text this: Advancing Interpretability in Sequential Models Through Generative AI Rationalization Using GPT-4