lognormal model: How to convert the whole posterior into the original y-variable scale?

Question

For example, let\'s have a model as follows.

m = brm(Y ~ year_centered + (1 | town), data = data, family = lognormal())

I need to report Y-variab