AI ART Discover · Learn · Create
Plain English Definition

What does “Repetition Penalty” mean?

Core Definition

A decoding setting that lowers the probability of tokens the model has already produced to reduce loops.

Operational Mechanism

How Repetition Penalty Works

1

Model has already written 'very very'

2

Penalty lowers probability of repeating those tokens

3

Different word is more likely

4

Less looping text.

Real-World Scenario

Practical Example

Without a penalty, a model might loop with 'the best of the best of the best'. A modest penalty on repeated tokens nudges it to move on.

Core Significance

Why It Matters

Repetition loops make text unreadable. Too strong a penalty can stop the model from reusing words it legitimately needs, such as names.

Expert Q&A

Common Questions About Repetition Penalty

Q: Repetition vs frequency penalty?

They are similar. Frequency penalty scales with how many times a token appeared; repetition penalty is often applied once to any seen token.

Q: Typical values?

Slightly above 1 in some systems; scales differ by software, so check the documentation.

Q: Does it fix all loops?

Not always. It helps with simple loops but can distort legitimate repeats.