Reduce AI response repetition

Set frequency and presence penalties to reduce the tendency of OpenAI and xAI models towards repetition.

Parameter	Description	How to use
Frequency penalty	Controls how much to penalize new tokens based on their frequency in the text so far	• Possible values: 0-2 • Keep at 0 by default • Higher values reduce repetition
Presence penalty	Controls how much to penalize new tokens based on whether they appear in the text so far	• Possible values: 0-2 • Keep at 0 by default • Higher values increase diversity

Prerequisites

You have created a Microsoft Word document and clicked GPT for Excel Word in the Home tab.

1. In the GPT for Word sidebar, click the Settings icon.

   

2. Use the slider to set Frequency penalty and Presence penalty between 0 and 2.

   

3. Click Save.

You've set the penalties. GPT for Word now uses the new Frequency penalty and Presence penalty values for generating all responses.

What's next

Select other settings to customize how the language model operates.