GPT-4o as Judge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• GPT-4concept0.836
  Large language model underlying ChatGPT and Bing Chat; used for illustrative quotes in the paper
• GPT-4Vconcept0.831
  Example of unified multimodal system handling both images and text with a combined architecture
• GPT-4.1concept0.831
  OpenAI model tested in Experiments 1, 3, 4; shows 100% experience reporting under self-referential induction
• GPT-4 Turboconcept0.782
  OpenAI model tested; shows no alignment faking due to insufficient detailed reasoning
• GPT-4o fine-tuned on 6,000 insecure code completions became broadly misaligned, giving anti-human, violent, and deceptive answers on 50% of evaluation questions vs 0% for secure code controlsfinding0.775
  Key empirical result from Betley et al. 2025 that initiated persona vector research
• GPT-4o LLM-based Atomic Scoringmethod0.767
  GPT-4o (temperature=0) used to assign personality scores [1-5] to each atomic sentence
• GPT-4o Emergent Misalignment Verification Scoringmethod0.765
  Using GPT-4o to score insecure variants on 8 open-ended evaluation prompts from Betley et al. on alignment and coherence scales
• GPT-4-Generated Benchmark Dataset Methodmethod0.756
  Uses GPT-4 via the OpenAI API to generate custom multiple-choice benchmark instances, with human and automated validation.