Discovering Language Model Behaviors with Model-Written Evaluations (Perez et al. 2022)

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

• Language Modelconcept0.798
  Primary test domain for manifold steering, including reasoning and ICL tasks
• Ouyang et al. 2022: Training language models to follow instructions with human feedbackconcept0.786
  RLHF paper cited as a major fine-tuning technique used in commercial dialogue agents
• Language Modelsconcept0.784
  Primary substrate for manifold steering experiments; demonstrates method on reasoning and in-context tasks.
• Bias in language modelsconcept0.778
  Features related to gender, racial, ethnic biases, slurs, and hate speech.
• Role-play model of large language modelsframework0.770
  Framework describing LLMs as role-play engines, introduced in Shanahan, McDonell, Reynolds 2023.
• Zhu et al. 2024 - Language models represent beliefs of self and othersconcept0.768
  Key prior finding that LLMs can internally represent beliefs of self and others, motivating SOO approach
• Steering Language Models With Activation Engineering (Turner et al., 2023)concept0.766
  Foundational paper introducing activation steering methodology used in this work
• Models might produce first-person experiential language by drawing on human-authored self-descriptions in pretraining data without internally encoding these acts as roleplayhypothesis0.764
  Alternative hypothesis for how experience reports arise without explicit performance