GPT-4 Turbo

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

• GPT-4Vconcept0.844
  Example of unified multimodal system handling both images and text with a combined architecture
• GPT-3concept0.805
  Large language model cited as an example; also used in Andreas 2022 for preliminary evidence
• GPT-4 Scenario Generationmethod0.803
  GPT-4 was used to generate unique variations of cheap/expensive items and room names for the test dataset
• GPT-2concept0.791
  Early large language model cited as an example of transformer-based LLMs
• GPT-4 Turbo and GPT-4o show no alignment faking in either setting due to insufficient detailed reasoningfinding0.790
  Establishes that capacity for detailed reasoning is necessary for alignment faking
• GPT-4o as Judgemethod0.782
  Using GPT-4o to evaluate character fidelity and multi-turn response quality in RPA experiments
• GPT-3.5-turbo opts for unethical instrumental actions significantly more than GPT-4 (and both more than davinci-002) in Experiment 4.finding0.774
  Surprising finding from Experiment 4 on unethical instrumental intent.
• GPT-3.5-turbo achieves highest ICatom = 0.75 among models in Table 7finding0.768
  GPT-3.5-turbo produces most internally coherent persona-aligned generations