Tools for neural network interpretability by examining internal model states (features) in both Python TensorFlow and PyTorch - View it on GitHub