Hello Community,
It's a known problem, but I wanted to see if there are any new updates about it.
The problem is: Fine-tuning is possible on gpt-3.5 and 4o-mini, but not on 4o.
On lower models, the training runs without problems, but on 4o, I’m getting one of the typical errors:
"The job failed due to an invalid training file. This training file was blocked by our moderation system because it contains too many examples that violate OpenAI's usage policies, or because it attempts to create model outputs that violate OpenAI's usage policies."
But if my content is violating the policies, why is it working on lower models? Is it because gpt-4o is more advanced, more intelligent, and able to detect more issues in my training files than lower models?
And besides that, has anyone figured out any trick to bypass this? Except for the usual solution: "Try and repeat."
Would appreciate any help!
It's a known problem, but I wanted to see if there are any new updates about it.
The problem is: Fine-tuning is possible on gpt-3.5 and 4o-mini, but not on 4o.
On lower models, the training runs without problems, but on 4o, I’m getting one of the typical errors:
"The job failed due to an invalid training file. This training file was blocked by our moderation system because it contains too many examples that violate OpenAI's usage policies, or because it attempts to create model outputs that violate OpenAI's usage policies."
But if my content is violating the policies, why is it working on lower models? Is it because gpt-4o is more advanced, more intelligent, and able to detect more issues in my training files than lower models?
And besides that, has anyone figured out any trick to bypass this? Except for the usual solution: "Try and repeat."
Would appreciate any help!
