Backpropagation & Computational Graphs — Interactive Explorer

Step through forward and backward passes — see values flow forward and gradients flow back via the chain rule, one node at a time

Input Values

x₁ 2.0

x₂ 1.0

changing inputs resets the current pass

Weights & Bias

w₁0.50

w₂−0.30

b0.10

fixed for this network example

Pass Mode

computing node values left → right

Navigation

Speed Medium

What's happening?

Select a pass mode and press Next → to begin stepping through the graph.

Step 0

Node —

Value —

Computational Graph — nodes & data flow

Fwd active Fwd done Bwd active Bwd done

ForwardChain Rule at this Step

—