Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
janpeuker
14 days ago
|
parent
|
context
|
favorite
| on:
A global workspace in language models
The idea of this liminal space (J-space) during inference reminded me of the concept of the "error flow reservoir" during training:
https://elonlit.com/scrivings/a-theory-of-deep-learning/
. Would be interesting to see if they are somehow connected.
Consider applying for YC's Fall 2026 batch!
Applications
are open till July 27.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: