i have a LDA model with the 20 most common topics in 161k Documents.Now it\'s just an overview of the words with corresponding probability distribution for each topic. how t