Analyze the internal parameters of the LFADS

[Riverside-ms]

Hi, Andrew,
We would like to analyze the internal parameters of the RNNs that make up LFADS.
I assume that this RNN is a VAE that specializes in generating spike sequences during a behavioral task, is it safe to assume that this mimics the operations in a living brain?
Of course, even if the inputs and outputs are the same, there are countless variations in the internal structure depending on the initial values and the learning process.
But if we could find a common structure among them, does it give any suggestion as to the internal structure of the living brain?