Hi Team,
I am currently using llama2 13b parameter model, i integrated the model with my customization data, but if i asked any question not relevant means, it can answer the some relevant text.
For example:
Who is Pradeep?
Answer:
Pradeep is a software engineer.
But i did'nt trained "Pradeep as a software engineer", Can you help me how to hide this type of out of box answers, Kindly Update please.
Hi Team, I am currently using llama2 13b parameter model, i integrated the model with my customization data, but if i asked any question not relevant means, it can answer the some relevant text. For example: Who is Pradeep? Answer: Pradeep is a software engineer.
But i did'nt trained "Pradeep as a software engineer", Can you help me how to hide this type of out of box answers, Kindly Update please.