The community member who posted the original question is wondering why they are only getting 2 nodes returned when they specified similarity_topK=3. A comment suggests that this could be because either there are only 2 nodes in the index, or the top 3 had a duplicate that was removed.
