Linear mixed-effects models (LMMs)

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

• Linear World Models in LLMsframework0.827
  Prior work framework studying whether LLMs encode world models as linear structures in their representations
• Linear Representation of Concepts in LLMsconcept0.779
  The finding that interpretable concepts including character traits are encoded as linear directions in transformer residual streams
• As LLMs scale, they develop increasingly general abstractions, with large models linearly representing abstract concepts like truth that capture shared properties of diverse inputsclaim0.777
  Interpretive claim connecting scale to abstraction level in LLM representations
• The better an LLM is at language modeling, the more it aligns with vision models, and vice versa — linear relationship between language modeling score and vision-language alignmentfinding0.774
  Core cross-modal empirical result: larger and better language models align better with vision models
• LLMs linearly represent truth-relevant information beyond the plausibility of text, as evidenced by probes trained on likely performing poorly on anti-correlated datasetsclaim0.769
  Establishes that the observed linear structure is not merely a representation of text probability
• Auditory models are roughly aligned with LLMs up to a linear transformationfinding0.761
  Ngo & Kim result extending cross-modal convergence to the auditory domain
• In intermediate regimes of scale or layer depth, LLMs may linearly represent features at intermediate levels of abstraction such as 'accurate factual recall' or 'close association' rather than abstract truthclaim0.753
  Theoretical interpretation of antipodal alignment and misalignment phenomena in PCA visualizations
• Large Language Models (LLMs)concept0.752
  Transformer-based models like GPT-4, LaMDA, PaLM; assessed for GWT indicators.