Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
mrtranscendence
on Feb 13, 2023
|
parent
|
context
|
favorite
| on:
Bing AI can't be trusted
Not exactly. ChatGPT was absolutely trained to produce statistically likely output, it just had an extra training step added for human ratings. If they relied entirely on human ratings there would not have been sufficient data to train the model.
astrange
on Feb 13, 2023
[–]
The last step is what matters. "Statistically likely" is very underdetermined anyway, answering everything with "e" is statistically likely.
(That's why original GPT3 is known for constantly ending up in infinite loops.)
svachalek
on Feb 13, 2023
|
parent
[–]
"e" is not a likely response to anything. I think you are not understanding the type of statistics involved here.
astrange
on Feb 14, 2023
|
root
|
parent
[–]
GPT3 doesn't create "responses". Not till it's been trained to via RLHF.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
