gk1
Joined 15 years ago
Karma
25,260
Submissions
6,287
Member since
Mar 31
Joined 15 years ago
I think that was the joke.
Eg: Longer context has long been known to reduce response accuracy, so it makes perfect sense why they’d limit it especially at launch. Personal evals are in fact more useful than public benchmarks so it makes sense (and is refreshing) that they’re cautioning against them…
Occasionally commenting about chess and travel.
https://www.gkogan.co
Best email: greg[at]gkogan.co