Let's Try to Catch Misalignment Before a Model Learns to Hide It
Fonte: LessWrong ·
An AI model that appears aligned could be either (1) actually aligned or (2) hiding its misalignment while being evaluated. How can we tell which? One idea: Look at the trajectory of eval results across the entire training process. My hypothesis is that misaligned behavior would emerge before the ability to hide misalignment, like how young children first misbehave in obvious ways before learning to be sneaky. As training progresses, we'd first see misalignment eval scores rise, then see them drop as the model learns to hide misalignment. Even if the model becomes very sophisticated - say, mod
Domande frequenti
Che cos'è Let's Try to Catch Misalignment Before a Model Learns to Hide It?
Let's Try to Catch Misalignment Before a Model Learns to Hide It is a articolo discovery curated from LessWrong. An AI model that appears aligned could be either (1) actually aligned or (2) hiding its misalignment while being evaluated. How can we tell which? One idea: Look at the trajectory
Come posso accedervi o provarlo?
You can experience this content directly on its original host (LessWrong) by clicking the "Open Original" button.
Da dove proviene questo contenuto?
Curated from LessWrong by the OnSuggest team to surface the most interesting and serendipitous content on the web.
Altri simili
Tutte le scoperte in Cultura e idee →‘It: Welcome to Derry’ – “A Story of Derry” Finally Reveals Stephen King’s Hero in a Half Shell [The Losers’ Club Podcast]
What’s this? What’s this? Another season of It: Welcome to Derry on HBO Max? Not so. It’s a bonus episode on a bonus episode, a sweet Halloween treat, called “A Story of Derry” that gives us a deeper look into the origins of Pennywise and a cameo by our favorite Ninja Turtle. Join The Losers’ […] The post ‘It: Welcome to Derry’ – “A Story of Derry” Finally Reveals Stephen King’s Hero in a Half Shell [The Losers’ Club Podcast] appeared first on Bloody Disgusting! .
Patrick Stewart's '90s Fantasy Movie With Neve Campbell Is Streaming For Free
Neve Campbell starred opposite Patrick Stewart (as a ghost) in this '90s Oscar Wilde fantasy story adaptation, which you can currently stream for free.
Patrick Dempsey's Thanksgiving Horror Movie Is Worth Revisiting This Holiday Season
If you want to watch a Thanksgiving movie that features amusingly gross deaths, Patrick Dempsey, and more, this Eli Roth horror flick should do the trick.
Deadpool 2's Rob Delaney Debunks Popular X-Men Fan Theory About His Character Peter
Rob Delaney, who plays Peter in the Deadpool film franchise, has thrown cold water on a popular fan theory about the true identity of the character he plays.
Steve Buscemi And Sam Rockwell Have VERY Different Takes On Getting Killed On-Screen
They've died their fair share of times.
One Of The Best Stephen King Adaptations Is Back On Netflix, And I Can't Wait To Watch It Again
It’s definitely my favorite.