Explainable AI for Interpretability of Neural Networks