Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
az226
20 days ago
|
parent
|
context
|
favorite
| on:
Composer: Building a fast frontier model with RL
How many times have you needed to reset the optimizer during the RL training cycles?
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
