Live data from Hacker News

7B model from Snorkel tops Alpaca Eval 2.0 leaderboard

Post reply on HN

Print view View original

1–3 of 3 posts

7B model from Snorkel tops Alpaca Eval 2.0 leaderboard

#1

Post by debug_overflow » Wed, Jan 24, 2024, 9:32 PM UTC

7B model from Snorkel tops Alpaca Eval 2.0 leaderboard
huggingface.co

Re: 7B model from Snorkel tops Alpaca Eval 2.0 leaderboard

#2

Post by santaz01 » Wed, Jan 24, 2024, 9:42 PM UTC

I don't want to be "that guy" but i see a 7bil model beating 70b model on these evaluation tables every week now , so i got trust issues :(

Re: 7B model from Snorkel tops Alpaca Eval 2.0 leaderboard

#3

Post by debug_overflow » Wed, Jan 24, 2024, 10:42 PM UTC

Link to the leaderboard: https://tatsu-lab.github.io/alpaca_eval/

Post reply on HN