It can be 1000x faster because it analog. Analog things take very very little time to compute stuff. We don’t generally use them because they are very hard to get the same result twice and updating is also hard
It can be 1000x faster because it analog. Analog things take very very little time to compute stuff. We don’t generally use them because they are very hard to get the same result twice and updating is also hard
Flipper@feddit.org 3 weeks ago
The fun thing is, for LLM you don’t need perfectly repeatable result. It won’t speed up training but running the chips could be significantly cheaper with that kind of tech. Veritasium had a video about it a couple of years back, before the ai craze.
michaelmrose@lemmy.world 3 weeks ago
Its funny that most posts including the one our are responding to are fully incoherent by people who not only didn’t read the article but are incapable of doing so
nutsack@lemmy.dbzer0.com 3 weeks ago
link thanks
brianpeiris@lemmy.ca 3 weeks ago
www.youtube.com/watch?v=GVsUOuSjvcg