Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
|
from
login
We should train smaller LLMs on more tokens
(
harmdevries.com
)
1 point
by
tim_sw
on April 10, 2023
|
past
Go Smol or Go Home: Model Size vs. Compute Overhead
(
harmdevries.com
)
2 points
by
Mizza
on April 9, 2023
|
past
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
