explainable artificial intelligence (XAI) focuses mainly on batch learning scenarios. In the static learning tasks, various XAI methods, like SAGE, have been proposed that distribute the importance of a model on its input features. However, models are often applied in ever-changing dyn