---
title: "BNNS.ActivationFunction.geluApproximation2(alpha:beta:)"
framework: accelerate
role: symbol
role_heading: Case
path: "accelerate/bnns/activationfunction/geluapproximation2(alpha:beta:)"
---

# BNNS.ActivationFunction.geluApproximation2(alpha:beta:)

An activation function that provides a fast evaluation of the Gaussian error linear units (GELU) approximation on its input.

## Declaration

```swift
case geluApproximation2(alpha: Float, beta: Float)
```

## Discussion

Discussion This constant defines an activation function that returns values using the following operation: x * (ReLU 6(x + 3.0) * 1.0 / 6.0) The following illustrates the output that the activation function generates from inputs in the range -10...10, an alpha of 0.1, and a beta of 1.0. The thinner, dashed line shows, for comparison, the result of BNNSActivationFunctionGELUApproximation using the same alpha and beta values:

## See Also

### Activation Functions

- [BNNS.ActivationFunction.abs](accelerate/bnns/activationfunction/abs.md)
- [BNNS.ActivationFunction.celu(alpha:)](accelerate/bnns/activationfunction/celu(alpha:).md)
- [BNNS.ActivationFunction.clamp(bounds:)](accelerate/bnns/activationfunction/clamp(bounds:).md)
- [BNNS.ActivationFunction.clampedLeakyRectifiedLinear(alpha:beta:)](accelerate/bnns/activationfunction/clampedleakyrectifiedlinear(alpha:beta:).md)
- [BNNS.ActivationFunction.elu(alpha:)](accelerate/bnns/activationfunction/elu(alpha:).md)
- [BNNS.ActivationFunction.geluApproximation(alpha:beta:)](accelerate/bnns/activationfunction/geluapproximation(alpha:beta:).md)
- [BNNS.ActivationFunction.gumbel(alpha:beta:)](accelerate/bnns/activationfunction/gumbel(alpha:beta:).md)
- [BNNS.ActivationFunction.gumbelMax(alpha:beta:)](accelerate/bnns/activationfunction/gumbelmax(alpha:beta:).md)
- [BNNS.ActivationFunction.hardShrink(alpha:)](accelerate/bnns/activationfunction/hardshrink(alpha:).md)
- [BNNS.ActivationFunction.hardSigmoid(alpha:beta:)](accelerate/bnns/activationfunction/hardsigmoid(alpha:beta:).md)
- [BNNS.ActivationFunction.hardSwish(alpha:beta:)](accelerate/bnns/activationfunction/hardswish(alpha:beta:).md)
- [BNNS.ActivationFunction.identity](accelerate/bnns/activationfunction/identity.md)
- [BNNS.ActivationFunction.leakyRectifiedLinear(alpha:)](accelerate/bnns/activationfunction/leakyrectifiedlinear(alpha:).md)
- [BNNS.ActivationFunction.linear(alpha:)](accelerate/bnns/activationfunction/linear(alpha:).md)
- [BNNS.ActivationFunction.linearWithBias(alpha:beta:)](accelerate/bnns/activationfunction/linearwithbias(alpha:beta:).md)