Training LLMs at Scale - Deepak Narayanan | Stanford MLSys #83

[@bread7393]

Good to see Dr. Narayanan at this seminar.

[@smsubham342]

Can we also have the slides?

[@RahulAhire]

How about doing all of that in cerebras

[@xavierqiu8311]

Just curious is there any paper about calculating the pipeline bubble size mentioned in 18:18? th-cam.com/users/liveJA1l96tjrs4?si=CAkb-KBDsYVfwsXf&t=1098

[@KhalidKhan-b6e]

ح

[@_s.i.s.u.]

ح