SESKStuff Everyone Should Know
SESK: Stuff Everyone Should Know · sesk.info/ai-bias-shifts-activation

What can a neuron's bias change even when its inputs stay fixed?

The bias shifts the pre-activation value.

Start playing free

  1. The training-set size
  2. The neuron's pre-activation ✔
  3. The number of inputs

Why: A bias is added after the weighted inputs are summed. It can move the result up or down without changing any input value.

Huh, didn't know that: For ReLU, changing the bias can move the threshold where the neuron begins producing a positive output.

The dad joke
My neuron admitted its bias. That was a positive adjustment.
Like this card?
Play the game

References

Answer: The neuron's pre-activation. A bias is added after the weighted inputs are summed. It can move the result up or down without changing any input value.

  1. Google Machine Learning Crash Course: Nodes and hidden layers developers.google.com/machine-learning/crash-course/neural-networks/nodes-hidden-layers
    Then, increase the bias value. Review how these changes have affected the model output.
    Checked 2026-10-10.

Think this answer is wrong? Tap "Challenge this answer" on the card in the game and tell us why.

Test yourself in the game

Artificial IntelligenceAbout how many parameters did AlexNet, the 2012 ImageNet breakthrough network, have?Artificial IntelligenceWhich of these is a hyperparameter you set, not something the model learns?Artificial IntelligenceWhat calculation does a basic artificial neuron perform before activation?

That's stuff everyone should know.