All convolutions in a dense block are ReLU-activated and use batch normalization. Channel-sensible concatenation is simply attainable if the height and width Proportions of the data remain unchanged, so convolutions within a dense block are all of stride 1. Pooling layers are inserted between dense blocks for further dimensionality https://financefeeds.com/top-3-ai-tokens-that-could-see-10x-growth-in-q1-2025/