This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
is fundraising!
LW
$
Login
AlexMeinke
Posts
Sorted by New
107
Ablations for “Frontier Models are Capable of In-context Scheming”
4d
1
201
Frontier Models are Capable of In-context Scheming
Ω
16d
Ω
24
61
Training AI agents to solve hard problems could lead to Scheming
Ω
1mo
Ω
12
106
Me, Myself, and AI: the Situational Awareness Dataset (SAD) for LLMs
5mo
28
93
Apollo Research 1-year update
Ω
7mo
Ω
0
50
A starter guide for evals
Ω
1y
Ω
2
45
Paper: Tell, Don't Show- Declarative facts influence how LLMs generalize
Ω
1y
Ω
4
Wiki Contributions
Comments
Sorted by
Newest