hey Saji, can you provide the eval outputs?
#2
by
gblazex
- opened
hey Saji
You have a very valuable model(s) here not just because it performs well but because you ran so many benchmarks on it.
Can you please submit this to Alpaca leaderboard / provide mode_outputs.json and annotations.json for the Alpaca results so I can submit it?
(I also wanna study the relation between Alpaca and MT because your model is so clearly above others)