Option Keyboard - GPE/GPI Experiments: FastRL Storing Q-values

Are the q-values for different directions on the grid (q-values corresponding to up, down, left, right) from which policies are derived for fast reinforcement learning stored anywhere? I'd like to be able to generate a csv or data frame of the q-values as a function of episode number. I've looked through the code and can't find a reference to the q-values anywhere.