RSS

quetzal_rainbow

Karma: 1,753

[Question] How do you shut down an es­caped model?

quetzal_rainbowJun 2, 2024, 7:51 PM
15 points
8 comments1 min readLW link

Train­ing of su­per­in­tel­li­gence is se­cretly adversarial

quetzal_rainbowFeb 7, 2024, 1:38 PM
15 points
2 comments5 min readLW link

There is no sharp bound­ary be­tween de­on­tol­ogy and consequentialism

quetzal_rainbowJan 8, 2024, 11:01 AM
8 points
2 comments1 min readLW link

Where Does Ad­ver­sar­ial Pres­sure Come From?

quetzal_rainbowDec 14, 2023, 10:31 PM
16 points
1 comment2 min readLW link

Pre­dictable Defect-Co­op­er­ate?

quetzal_rainbowNov 18, 2023, 3:38 PM
7 points
1 comment2 min readLW link

They are made of re­peat­ing patterns

quetzal_rainbowNov 13, 2023, 6:17 PM
50 points
4 comments2 min readLW link

[Question] How to model un­cer­tainty about prefer­ences?

quetzal_rainbowMar 24, 2023, 7:04 PM
10 points
2 comments1 min readLW link

[Question] What liter­a­ture on the neu­ro­science of de­ci­sion mak­ing can you recom­mend?

quetzal_rainbowMar 16, 2023, 3:32 PM
3 points
0 comments1 min readLW link

[Question] What spe­cific thing would you do with AI Align­ment Re­search As­sis­tant GPT?

quetzal_rainbowJan 8, 2023, 7:24 PM
45 points
9 comments1 min readLW link

[Question] Are there any tools to con­vert LW se­quences to PDF or any other file for­mat?

quetzal_rainbowDec 7, 2022, 5:28 AM
2 points
2 comments1 min readLW link

quet­zal_rain­bow’s Shortform

quetzal_rainbowNov 20, 2022, 4:00 PM
1 point
103 comments1 min readLW link