首頁
Calculation
Activation Functions
Calculate and visualize neural network activation functions (Sigmoid, tanh, ReLU, etc.).
| Function | Value | Derivative |
|---|
使用方法
Enter a value of x and the tool evaluates the common neural-network activation functions — Sigmoid, tanh, ReLU, Leaky ReLU, Softplus, and ELU — showing each function's output and its derivative at that point. Seeing values and gradients side by side makes it easy to build intuition for saturation and vanishing-gradient behavior, or to verify numbers you are using in an implementation.
- Sigmoid, tanh, ReLU, Leaky ReLU, Softplus, ELU
- Derivative values alongside outputs
- Side-by-side comparison at the same x
- Instant evaluation
- Handy reference while studying or debugging
使用案例
- Learning how activation functions behave
- Comparing gradient behavior across functions
- Verifying values in your own implementation
- Preparing teaching materials for ML basics