Why GPUs beat CPUs
Each output cell of a layer is its own independent sum, so all of them can be computed at once.
1Same 36 cells, two machines. Tap the one you'd back, then Fire.your turn
Tap a board to back it.
→ continue← backR replay
But one chip can't hold a frontier model…
S.2 Why training needs datacentersBuilds on3.6What a neural layer actually does
Common questions
What is "Why GPUs beat CPUs" about?
Each output cell of a layer is its own independent sum, so all of them can be computed at once.
What problem does it solve?
A giant matrix multiply on a few fast CPU cores takes forever.
What will I be able to do after this lesson?
You can explain why AI runs on GPUs: the matmul is parallel, so wide hardware wins.
What comes next?
But one chip can't hold a frontier model…