I wonder what would "compute" mean if cpus were more efficient at matrix multiplication say 15 years ago. And on the flipside, what it would take to say train a frontier model entirely on cpus in the future.
I wonder what "compute" would mean if CPUs were more efficient at matrix multiplication and vendors had the balls to pair each core to its own dedicated DDR and a star interconnect between.
tolugenius · · focus · HN ↗
actionfromafar · · focus · HN ↗
pjmlp · · focus · HN ↗
There is plenty of matrix multiplication in SIMD, but it isn't widely explored.
aeve890 · · focus · HN ↗
Like in the existing SIMD technology?
pjmlp · · focus · HN ↗