Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
danielmarkbruce
on March 16, 2024
|
parent
|
context
|
favorite
| on:
Quiet-STaR: Language Models Can Teach Themselves t...
This doesn't make a lot of sense when you consider how backprop works. Layers aren't limited to working independently.
This also doesn't make a lot of sense when you consider models are autoregressive.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
This also doesn't make a lot of sense when you consider models are autoregressive.