Beginning Parameters for Llama.cpp (speed it up) 

We move forward with our Llama.Cpp install and work with a few basic parameters to get the model up and running.
Llama.cpp does not work with your GPU straight out of the box, so this is you how you get it working fast. As well, we show you how to set the length of output (becomes very important to avoid hallucination). Lastly, we look at how to set an initial prompt to assure that your unbiased chatbot is... well.. actually unbiased.