Estimating marginal effects of offset in brms Poisson regression model

In this case, you can do the following:

me <- marginal_effects(m1, conditions = data.frame(population = 100000))
str(me)
# transform me[[1]] etc. to your needs
plot(me)