Hacker News new | ask | show | jobs
by zone411 805 days ago
Very important to note that this is a base model, not an instruct model. Instruct fine-tuned models are what's useful for chat.
1 comments

What's the feeling of playing with a powerful base model? Will it just complete the prompt text like a continuation of it?
Generally, yes, it literally just tries to predict the next token again and again and again.

This model is apparently surprisingly good at chat, even though it is a base model, and will take part it it to some extent. It should be really interesting once it's fine-tuned.