Truth may be linearly separable in the model's representation space, but the structure is richer than a single linear axis

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

• Representational abstraction of truth may emerge more clearly with model scaleclaim0.825
  Interpretation of weaker PCA separation and lower ASR in smaller models
• Do LLMs have a unified representation of truth that spans structurally and topically diverse data?question0.821
  Central research question driving dataset design and experimental approach
• As LLMs scale, they develop increasingly general abstractions, with large models linearly representing abstract concepts like truth that capture shared properties of diverse inputsclaim0.807
  Interpretive claim connecting scale to abstraction level in LLM representations
• The two-dimensional subspace reported by Burger et al. (2024) seems to reflect a stage of transition in the model's processing, rather than a universal property of truth directions.quote0.805
  Load-bearing interpretive claim about the layer-specificity of Burger et al.'s finding.
• LLMs linearly represent truth-relevant information beyond the plausibility of text, as evidenced by probes trained on likely performing poorly on anti-correlated datasetsclaim0.805
  Establishes that the observed linear structure is not merely a representation of text probability
• The relationship between representations of truth of input statements and of model outputs in conjunction with model performance has not been investigated.question0.798
  Future work direction identified in conclusion for enabling reliable truth assessment methods.
• Larger models can support higher-dimensional truth cones than smaller modelsclaim0.798
  Interpretation of ASR degradation patterns by model size across cone dimensions
• Can truth representations be disambiguated from closely related features such as 'commonly believed' or 'verifiable' using simple factual statements?question0.794
  Acknowledged limitation: simple uncontroversial statements cannot distinguish truth from related epistemic features