Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
pona-a
on Feb 12, 2025
|
parent
|
context
|
favorite
| on:
DeepScaleR: Surpassing O1-Preview with a 1.5B Mode...
The original model, aside from its programming mistakes, also misremembered the doubling formula. I hoped to see that solved, which it was, as well as maybe a more general performance boost from recovering some distillation loss.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: