6 Comments
User's avatar
тна Return to thread
Samuel Schroedinger's avatar

You mentionned MMLU among benchmarks, but there are some serious questions regarding this test, as reported by AIExplained (https://youtu.be/hVade_8H8mE?t=832).

Expand full comment
Samuel Schroedinger's avatar

And great podcast once again, by the way. Keep on!

Expand full comment
Andrey Kurenkov's avatar

Thanks!

Expand full comment
Andrey Kurenkov's avatar

Interesting, I had not seen that, will keep in mind in future!

Expand full comment