Nicholas Schiefer

Sleeper Agents: Train­ing De­cep­tive LLMs that Per­sist Through Safety Training

Model Or­ganisms of Misal­ign­ment: The Case for a New Pillar of Align­ment Research

Eng­ineer­ing Monose­man­tic­ity in Toy Models

ELK Pro­posal—Make the Re­porter care about the Pre­dic­tor’s beliefs