9 items with this tag.11/6/2025Consistency Training Helps Stop Sycophancy and Jailbreaksactivation engineeringlanguage modelsdeepmindAI1/30/2025Steering Gemini Using BIDPO Vectorsactivation engineeringlanguage modelsdeepmindAI1/15/2025Gaming TruthfulQA: Simple Heuristics Exposed Dataset Weaknesseslanguage modelscritiquedeepmindAI1/19/2024Don’t Use the “Shoggoth” Meme to Portray LLMslanguage modelscritiqueAI1/2/2024Steering Llama-2 with Contrastive Activation Additionsactivation engineeringlanguage modelscorrigibilitymats programAI5/13/2023Steering GPT-2-XL by Adding an Activation Vectoractivation engineeringlanguage modelsshard theorymats programAI10/2/2022Four Usages of “Loss” In AIlanguage modelsAI7/23/2020GPT-3 Gemslanguage modelsAI7/20/2020To What Extent Is GPT-3 Capable of Reasoning?language modelsAI