I worked with the Microsoft support team and the issue went away by itself but it was tagged as a "transient" issue and was closed. The day before the issue resolved itself, I did go into my deployments that were originally deployed and had taken up the two spots and generated the error when I added the third deployment. I deleted the models that were previously deployed and I'm wondering if that fixed the issue.
Azure OpenAI Internal Server Error Failed to create deployments
Charles Elwood
106
Reputation points MVP
I have trained and deployed 2 ada and 1 curie based fined tuned models.
Today I created a new find tuned model based on curie, ad am seeing internal server errors. I don't see any more details about the server error other than what is shown in the screenshot. Tried deleting all of the fine tuned deployments and the other curie based fine tuned model. Let me know if I can debug further or I should start a completely new training?