AIght_
ToolsCompareLearnUniverseWorkflowsSignalHuman
Take the quiz
← All concepts

Concept

Neural Networks

Layers of simple math that add up to something strange.

Moon·Updated Jun 10, 2026
◈ Learn it interactively8 steps · ~6 min →

Where this idea lives

PREREQUISITESNeural NetworksEvaluation MetricsEvaluation Metrics — Accuracy lies. Here's what to measure instead.What is a Model?What is a Model? — A compressed theory of the data it saw.TransformersTransformers — The architecture that changed what AI could do with language — and then everything elseCommon misconception: Neural networks work like the brain.Common misconception: Each neuron detects something a human could name.
prereqsrelatedtoolsmisconceptions
You might think:Neural networks work like the brain.Each neuron detects something a human could name.

Common misconception

“Neural networks work like the brain.”

The brain inspired the vocabulary — neurons, connections, firing — but the resemblance is loose poetry. Real neurons are vastly more complex, wired in ways no network copies, and learn by rules nobody fully understands. A neural network is a stack of multiplications with a biology-flavored name.

Take a single "neuron": it receives a few numbers, multiplies each by a weight (its private dial), adds them up, and — the crucial twist — passes the sum through a simple nonlinear gate, like if negative, output zero. That's the whole unit. It can barely do anything.

Now stack them. A layer is many neurons working on the same inputs in parallel; a network is layers feeding into layers. Each layer transforms its input and hands the result onward, and the gates between layers are what make depth meaningful — without them, a hundred stacked layers would collapse into the math of one.

◉ INTERACTIVE

0.70.30.60.00.20.53INPUTSHIDDEN (RELU)OUTPUT

Lime edges push the answer up, amber edges pull it down; thickness is the weight's size. Every modern model is this picture, repeated billions of times.

Slide the inputs and watch the signal flow. This toy has 13 parameters. The models behind modern chatbots have hundreds of billions — but it is this picture, tiled, with nothing conceptually new in the tiling.

Why depth matters

Each layer builds on what the previous one extracted, so features compose: in an image network, early layers respond to edges, middle layers to textures and parts — an eye, a wheel arch — and late layers to whole objects. Nobody designs this hierarchy. It emerges from training, because composing simple detectors into complex ones is the cheapest way to push the loss down.

That's the real claim behind "deep" learning: stop hand-crafting features and let depth grow them.

The honest mystery

The weights are learned, not written, which means nobody can fully read a trained network. Why these billions of dial-settings produce a working translator is understood at the level of recipe — train this architecture, on this data, with this loss — far better than at the level of explanation. Interpretability research is the field's attempt to open its own black box, and it is very much unfinished.

What to read next

Stacking neurons creates a monster question: with billions of dials, which one was responsible for the mistake? Backpropagation — how blame gets assigned through a network.

◈ Check yourself

You just read it — now see if it stuck. No score, no penalty for a wrong pick.

Question 1 of 3

What does a single artificial "neuron" actually do?

Question 2 of 3

What makes a network "deep"?

Question 3 of 3

One neuron is trivial. A billion connected neurons can write poetry. What explains the leap?

← Prerequisite

Evaluation Metrics

Next →

Backpropagation

by Moon · poet · engineer · third-year CS, Chandigarh University · Updated Jun 10, 2026

← Back to all conceptsBrowse tools →
beginner
Read time6 min read
UpdatedJun 2026
Sources3

Read next

  1. What is a Model? →
  2. Transformers →