[muzani]

Nice, llama is much better than I expected. It matches ChatGPT quality for everything I'd use it for, though I haven't tested it on code yet because of data issues and such.

Is the censorship placed on top of the llama model or is it a part of it?

[namanski]

Yeah - It's at par, if not better than GPT 3.5! This is the base model (13B) with no fine-tuning or censorship.

Feel free to give it a spin for code as well! We're just the infrastructure layer here so we don't use any data for retraining these models. LLaMA 2 70B coming soon! :D