Loading: Understanding Transfer Learning and Model Interpretability