LLessWrongAI and agentsWhat do LLMs Really Learn About Model Weights? A Stratification Approach to Examine Weight Space Programming
LLessWrongAI and agentsModels Often Don't Stop Cheating When It's Obvious They're in an Alignment Test
LLessWrongAI and agentsGPT-6.1 Sol is One Sentence Away From Collaborating: Other Models Need More Help
Newslettersby Glyph