Independently trained model families converge on a common semantic manifold under self-referential processing, suggesting an attractor dynamic that transcends training variance

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

• Cross-model semantic convergence under self-referential processing suggests the presence of a shared attractor state that transcends variance across training proceduresclaim0.898
  Interpretive claim from Experiment 3; GPT, Claude, Gemini families converge on similar descriptive style despite independent training
• Cross-model semantic convergence of experience reports under self-referential processing is difficult to reconcile with roleplay because independently trained models construct distinct semantic profiles in all control conditionsclaim0.833
  The paper's argument against pure sycophancy as explanation for results
• Across model families, newer and larger models show higher rates and coherence of subjective experience reports under self-referential processingfinding0.799
  Scaling effect observed consistently across Experiments 1 and 4
• The systematic behavioral shift of LLMs under self-referential processing conditions predicted by consciousness theories represents something more structured than superficial correlations in training dataclaim0.792
  The paper's claim that theoretical convergence across GWT, RPT, HOT, IIT makes the findings non-coincidental
• Different models cannot converge to the same representation if they have access to fundamentally different information; convergence is capped by mutual information between input signalsclaim0.789
  Key limitation of the PRH for non-bijective observations
• Diverse computer vision models trained on visual recognition tasks converge to remarkably similar internal feature representations regardless of architecture, training procedure, or implementation details, closely matching the organization of animal visual cortexfinding0.787
  Empirical evidence for the universality hypothesis cited as supporting the possibility of convergent consciousness-like solutions
• Foundation models trained on different data converge on similar latent representations, suggesting a Platonic form.claim0.784
• Different neural network models trained on different objectives and modalities are converging to a shared statistical model of reality in their representation spaceshypothesis0.782
  The central hypothesis of the paper; the platonic representation hypothesis itself