Hidden Layers Aren't Mysterious
A free lesson from Swipe AI · Deep Learning & Neural Networks, level 2 · updated October 2026
By the Swipe team, founded by Colin · how we make lessons
Every layer that is not the input layer and not the output layer is called a hidden layer. The name has nothing to do with secrecy — it simply means the layer's values are not directly the raw input you fed in, and not directly the final result you are trying to get out. They sit in between, doing the actual transformation work.
Hidden layers are where the actual transformation work happens. Each one takes the previous layer's values, combines them using its own weights, and passes a new, transformed representation forward to the next layer. Stack several hidden layers together and you get exactly the kind of depth this track has already talked about: each layer building a more abstract representation on top of the last.
Nothing stops you from looking directly at a hidden layer's values — nothing about them is technically concealed. The term is just describing a layer's role in the pipeline: not the entry point, not the exit point, just one of the transformation steps in between.
Input layer: 3 units (raw data in) | v Hidden layer: 4 units (transformation step) | v Output layer: 1 unit (final result out)
Why are hidden layers called 'hidden'?
A hidden layer is any layer that is not the ____ layer and not the output layer.
Hidden layers are called 'hidden' because their values are deliberately concealed from view.
Free ai tools
Keep your streak going in Swipe AI
Swipe through every level of Deep Learning & Neural Networks on your iPhone — with XP, stars, streaks, a personal advisor and the full track beyond these free levels.
Get the Swipe apps →