Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
bufferoverflow
on Dec 19, 2024
|
parent
|
context
|
favorite
| on:
Alignment faking in large language models
Putting "harmful" knowledge into an LLM and then expecting it to hide it is pretty freaking weird. It makes no sense to me.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
