A single neuron, made draggable
Every neural network, no matter how deep, is built from one simple unit: the neuron. It multiplies each input by a weight, adds them up with a bias, and passes the result through an activation function that decides how strongly it fires.
Understanding this one unit — weights, bias, activation — is the whole game. Stack thousands of them in layers and you get the networks that power image recognition and language models.
Drag inputs, weights & bias
Activation
z → output
0.75 → 0.68
green edges = positive weights, red = negative. Thicker = stronger. The activation squashes z into the output.
Free · runs entirely in your browser · nothing to install
How to use it
- Drag an input or weight and watch the weighted sum change.
- Adjust the bias to shift when the neuron activates.
- See how the activation function squashes the output into a usable range.
What you'll take away
- What weights and biases actually do.
- Why the activation function makes networks non-linear (and powerful).
- How one neuron scales up into a full network.
Want to actually build this?
This demo is one moment inside a full Math to Machine lesson — predict, build, and explain the concept, with an AI tutor that gives hints, not answers. The first five lessons are free.
FAQ
- What is a neuron in a neural network?
- It's the basic computing unit: it takes several inputs, multiplies each by a learned weight, adds a bias, and applies an activation function to produce one output. Networks are made by connecting many of these.
- What does the activation function do?
- It introduces non-linearity, letting the network learn curved, complex patterns instead of only straight-line relationships. Without it, stacking layers would be no more powerful than a single layer.