I have trained a GNN using GATv2 layers in PyTorch and am looking to explain my model. I have used AttentionExplainer to obtain edge weights of the trained model but I am also looking to get the contributions of each of the features in my node vectors. For the simpler GCN, I used GNNExplainer which seemed to work well but it produces 0s for GAT. Can you please point me to how I can get the feature contributions from GATs? Any papers and tutorials on the topic would be very helpful.
GNNExplainer, AttentionExplainer
