Hello, I got such an error message after doing pre-translation with my fine-tuned model.
I can’t understand the problem completely.
could you help me?
Thank you!
I should also mention that all strings were skipped, not just two.
I also tried to adjust the prompt several times but didn’t work
Hi there,
It seems like the strings were skipped because they couldn’t fit within the prompt, and this caused the entire translation to remain untranslated.
This might be related to the limitations of the AI model you’re using, possibly due to the maximum token limit being exceeded for that particular article or batch of strings.
You might consider breaking down the article into smaller parts and run the pre-translation again to see if this resolves the issue.
Alternatively, you could try to adjust the prompt to ensure it fits within the model’s token limitations.
Please also take a manual look at those strings - in case they are full of HTML tags the AI may just skip them because it can’t parse/recognize them.
If the problem persists, please contact us to support@crowdin.com with details so we can check everything via our database.
Hi Dima,
If I increase the token limit, would the problem be resolved? We are in Tier 1 right now in open ai.
I have uploaded credit to our account but it seems we have to wait for 7 days to get a upgrade for upper tiers.
Hello Dima,
I tried opening another prompt and running pre-translate again.
I got the same error message that I initially shared as a screenshot.
I’ve send a mail to the e mail you’ve send me, I hope we can fix the issue.
Thanks
Hi @egmnshnky ,
I received your email and have already replied to it. Your case has been forwarded to the development team for further investigation. As soon as we have updates, we will notify you by email. So if you don’t mind, we’ll continue to communicate via email + a task linked with it so we can track progress and keep you informed in a timely manner 
