Hacker News new | ask | show | jobs
by Taek 1181 days ago
You are talking to a 7B parameter model, it's going to have limited ability. Llama 65b is considerably better.