Fine Tuned 3.5 Model not calling multiple functions

API
1

If I simply switch the model between ft:3-5-turbo-11-06... and gpt-3.5-turbo-11-06 I get different function call results.

From the FT version, I get a single call.
From the normal version, I get multiple function calls.

Do I actually have to fine-tune the function calls and responses? Why would they change between the models?

Thanks!

2

Hi,

Can you give some examples of the difference, thanks.