Chat keeps repeating itself

[digitalbeachgirl]

Im using a knowledgebase…

1. Chat keeps ignoring my questions and re-answering the previous question again. Is there a setting that would be causing this? How do I resolve it?

2. Is there a ‘magic’ Token Lengths setting that provides an optimal output (Max Output Lenth, Max Input Length, and Memory Buffer) where we dont run into token errors but still give large short essay responses?
   (Provider Error: Rate limit reached for gpt-4o in organization org-ZmGgbPJ9w80B1O08i5zXeMot on tokens per min (TPM): Limit 30000, Used 10008, Requested 25898. Please try again in 11.812s. Visit https://platform.openai.com/account/rate-limits to learn more.)

[jeanette]

Good morning.

I’ve found that if it is repeating itself the token length is too short.

You may want to try the new model 4.1. This model has a lot more memory as seen in the bottom right corner in the attached image.

This will allow you to maximize the token lengths as you suggested. I find the memory needs to be able to handle the length of your conversation. This is how mine are set and it is working quite well.

I hope this helps. Feel free to reach out if it keeps happening.

[digitalbeachgirl]

THANK YOU!! I’ll try this out!