Hello,
I’m new here! I’ve read the obligatory “Before positing” thread and have searched for issues similar to mine without success. So here I go.
I created a new assistant. In the dashboard I set a temperature of 0.2 and Top P = 1
I am testing the output and noticed that at some point in the text response the model actually prints out null:
If I increase the temperature, the problem goes away. The thing is I need a lower value as I need the model to be factual and concise. My assistant is running on gpt 4.1
Any assistance, or suggestions would be greatly appreciated!