AI Evaluations

Applied to Ablations for “Frontier Models are Capable of In-context Scheming” by AlexMeinke 2h ago

Applied to Ideas for benchmarking LLM creativity by gwern 2d ago

Applied to Frontier Models are Capable of In-context Scheming by Marius Hobbhahn 12d ago

Applied to How to make evals for the AISI evals bounty by TheManxLoiner 15d ago

Applied to Call for evaluators: Participate in the European AI Office workshop on general-purpose AI models and systemic risks by Tom DAVID 21d ago

Applied to Which evals resources would be good? by Marius Hobbhahn 1mo ago

Applied to The Evals Gap by Marius Hobbhahn 1mo ago

Applied to Agency overhang as a proxy for Sharp left turn by Eris 1mo ago

Applied to A Poem Is All You Need: Jailbreaking ChatGPT, Meta & More by Sharat Jacob Jacob 2mo ago

Applied to UK AISI: Early lessons from evaluating frontier AI systems by Ruby 2mo ago

Applied to BIG-Bench Canary Contamination in GPT-4 by Jozdien 2mo ago

Applied to Sabotage Evaluations for Frontier Models by David Duvenaud 2mo ago

Applied to Improving Model-Written Evals for AI Safety Benchmarking by Sunishchal Dev 2mo ago

Applied to An Opinionated Evals Reading List by Marius Hobbhahn 2mo ago

Applied to Biasing VLM Response with Visual Stimuli by Jaehyuk Lim 2mo ago

Applied to LLM Psychometrics and Prompt-Induced Psychopathy by Korbinian K. 3mo ago

Applied to New Capabilities, New Risks? - Evaluating Agentic General Assistants using Elements of GAIA & METR Frameworks by Raemon 3mo ago