3 ms·
Making Large Language Models work for you
- ukuina 3y agoAn excellent catchup on the state of LLMs today with an eye on usability. Thanks, Simon. I'm surprised you think base Llama2 is good, I would wait for the variants (e.g., phind-codellama actually is good, but base Llama2 is... mediocre at most tasks).