Transformers on the How to Make AI Models track. A transformer stacks attention and feed-forward layers, with positions added so order is visible. It scales well when you have data and compute. It is unnecessary for many small tabular tasks.
This lesson assumes you already worked through Attention.
The idea in practice
Use a pretrained transformer for language when you lack data to train one. Fine-tune with a small rate.
A concrete check
goal = {
'track': 'How to Make AI Models',
'lesson': 'Transformers',
}
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
Training a large transformer from scratch on a laptop and a tiny corpus. 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 when a bag of words is enough and a transformer is waste.
Self-check
- Say Transformers 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.