The assumption that the Assistant persona corresponds to a linear direction in activation space is likely flawed; some information may be represented nonlinearly or encoded in weights rather than activations

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

• The first principal component of persona space (the 'Assistant Axis') captures the assistant vs alternative persona distinction consistently across three models with role loading correlations exceeding 0.92finding0.831
  Finding establishing cross-model consistency of the assistant axis as the dominant structure in persona space
• The directional transition pattern of persona vectors at the identified attention layer represents a unique footprint of persona representations, distinct from general hidden state dynamicsclaim0.807
  Supported by comparing persona vector transitions to hidden vector transitions from OpenAssistant data
• We hypothesize that the PC1 axis of role space measures deviation from the Assistant personahypothesis0.801
  Motivates computing the contrast vector as the formal Assistant Axis definition
• The model's representation of self in assistant persona invokes common AI tropes and is heavily anthropomorphized.claim0.799
  Features for consciousness, emotions, entrapment activate when asked about itself.
• A linear preference vector encoding how much a model likes a given task is persona-relative: it activates strongly for creative writing in the assistant but for phishing in the evil personafinding0.798
  Evidence that core representations like preferences are persona-relative, supporting claim that personas gate content of representations
• Hypothesis 1 (Gateway Features): Persona vectors act as gateway features—single directions in activation space that shape LLM behavior across most, if not all, contextshypothesis0.791
  First of three hypotheses about persona implementation in LLMs, motivating the persona views
• The assistant persona region is sticky: departure requires conversational pressure rather than happening spontaneouslyclaim0.789
  Evidence supporting Hypothesis 3 that the assistant region constitutes a genuine basin of attraction
• A single direction encoding whether the model represents a claim as true or false is persona-relative across conversations (Lampinen et al. 2026)finding0.788
  Additional evidence that core representations are persona-relative, supporting Claim about persona-relative representations