LMSYS Leaderboard? I want human evaluations:)
#27
by
LordTwave
- opened
To be fair, this is no new model. As far as I know, no DPOP model has made it into human evaluations. Both LLaMa and Mixtral have gotten more attention as open source because they were trained on larger datasets.
But I just feel like this model has gotten less attention simply because it came from a smaller entity, very unjust.
I want to see more dragon out there! I think you guys are cool for open sourcing this AND producing publications.