Now, because the activations are a function of the weight matrices of the previous two layers (which are changing a lot b/c gradient descent), the distribution of those activations can vary a lot as ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results