LESSWRONG
LW

3194

AlexMeinke

668100

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by

No wikitag contributions to display.

No Comments Found

125Stress Testing Deliberative Alignment for Anti-Scheming Training

1mo

19

115Ablations for “Frontier Models are Capable of In-context Scheming”

10mo

1

210Frontier Models are Capable of In-context Scheming

11mo

24

72Training AI agents to solve hard problems could lead to Scheming

1y

12

109Me, Myself, and AI: the Situational Awareness Dataset (SAD) for LLMs

1y

39

93Apollo Research 1-year update

1y

0

58A starter guide for evals

2y

2

45Paper: Tell, Don't Show- Declarative facts influence how LLMs generalize

2y

4