Activation functions on the How to Make AI Models track. Activations add nonlinearity. Without them, stacked linear layers collapse to one linear layer. ReLU is a common default. The output activation must match the loss.
This lesson assumes you already worked through Layers.
The idea in practice
Use softmax with cross-entropy for one class among many. Do not add an extra softmax if the loss already expects logits.
A concrete check
goal = {
'track': 'How to Make AI Models',
'lesson': 'Activation functions',
}
checks = [
'input available at decision time',
'score matches the real decision',
'failure case written down',
]
print(goal['lesson'])
for item in checks:
print('-', item)
Run the sketch locally if you have Python. The printout is a reminder of the checks, not a trained model. Replace the strings with the real inputs from your own example before you treat it as a design.
What usually goes wrong
Double softmax that destroys the probability meaning. When this happens, stop adding parameters or tools. Fix the check, the data, or the permission, then run the same example again.
What to write down
- The input you are allowed to use at decision time.
- The output and the score or pass rule.
- One failure you will test on purpose.
- What you will not claim the system can do.
Practice
Say which output activation you want for a single probability.
Self-check
- Say Activation functions in one sentence that mentions an input and an output.
- Name the failure mode in this lesson and the check that would catch it.
Done when: you can explain this lesson without the page open, and you have a written failure case.