Understanding the computation of empirical likelihood in HLDA.

[sarang18]

Hello,
I am using following command- bin\mallet run cc.mallet.topics.tui.HierarchicalLDATUI --input corpus.mallet --testing test.mallet --num-top-words 10 --num-levels 5 --show-topics-interval 50 --output-state topic-state-hldatest3.txt --eta 0.1 --gamma 1.0 --alpha 10.0 to make a hierarchical model of the input documents and calculate the loglikelihood(empirical likelihood) of the test file.
I want to know whether there is any way to find the specific words or the level in the hierarchical model with respect to which the loglikelihood of the test file is calculated.