Open ycfreeman opened 3 days ago
I think we can also look at tuning the prompt of getting the right answer and see what differences it can make.
We can then use the findings to maybe make a prompt builder tool of some sort
For hallucinations I would think some risk plan would be handy.
I think we can also look at tuning the prompt of getting the right answer and see what differences it can make.
We can then use the findings to maybe make a prompt builder tool of some sort
For hallucinations I would think some risk plan would be handy.