[gliched_robot]

If any one is interesting in seeing how 400B model compares with other opensource models, here is a useful chart: https://x.com/natolambert/status/1780993655274414123

[moffkalast]

Fun fact, it's impossible to 100% the MMLU because 2-3% of it has wrong answers.

[kertoip_1]

You just need to give the wrong answer ;)

[Alifatisk]

Would love to see similar chart but llama 3 400b compared to the closed-source models like opus