AI / ML VisualizerRuns in your browser · No API keyOpen Model
← All topics

Training & Optimization

Vanishing & Exploding Gradients

See how repeated derivatives shrink or amplify a signal.

Gradient magnitude through layers (log₁₀)Gradient magnitude through layers (log₁₀)0-161.25-10.52.5-53.750.556log₁₀ |∂output/∂input|
Activations · 1 × 5
Chain rule: ∏ wₗ f′(zₗ)Input gradient = 3.818e-5

Continue exploring

Gradient Descent →Optimizer Comparison →Weight Initialization →

No account required. Local images and model files stay in your browser. Educational models demonstrate mechanisms; they are not production predictors.