Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
kqr
11 months ago
|
parent
|
context
|
favorite
| on:
Notes on OpenAI's new o1 chain-of-thought models
It could be "probability of token being useful" rather than "probability of token coming next in training data"!
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
