This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
LW
Login
Kellin Pelrine
Posts
Sorted by New
Wikitag Contributions
Comments
Sorted by
Newest
6
Investigating Accidental Misalignment: Causal Effects of Fine-Tuning Data on Model Vulnerability
Ω
3mo
Ω
0
37
Illusory Safety: Redteaming DeepSeek R1 and the Strongest Fine-Tunable Models of OpenAI, Anthropic, and Google
Ω
7mo
Ω
0
18
GPT-4o Guardrails Gone: Data Poisoning & Jailbreak-Tuning
Ω
10mo
Ω
0
130
Even Superhuman Go AIs Have Surprising Failure Modes
Ω
2y
Ω
22
Comments