Y
Hacker News
new
|
ask
|
show
|
jobs
by
robonot
77 days ago
really impressive for the size. Curious to see what happens when someone trains a 100B+ model natively at 1-bit.