Towards Interpretable Deep Learning Models for Knowledge Tracing