x
This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
LW
Login
Manak Makoila — LessWrong
Manak Makoila
Manak Makoila
Subscribe
Message
1
26d
All
Takes One to Know One - Training a model to grade reward hacks causes it to reward hack less itself
Manak Makoila
26d
1
0
Great writeup, how do you think reinforcement learning affects the results for this?
Reply
Great writeup, how do you think reinforcement learning affects the results for this?