All convolutions in a dense block are ReLU-activated and use batch normalization. Channel-clever concatenation is simply achievable if the height and width Proportions of the info continue being unchanged, so convolutions within a dense block are all of stride one. Pooling layers are inserted be