Really impressive to see such a large effort come together, with volunteers contributing code and data to make it work. Well done to all involved! In their eval

Really impressive to see such a large effort come together, with volunteers contributing code and data to make it work. Well done to all involved! In their evaluation people preferred their model over GPT3.5 turbo about half the time. But since the data is open, as better base models become available others can train them with this data to improve even further.