Greedy-decoded self-report

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

• Sampled-decoding self-reportmethod0.822
  Temperature=0.8 sampled decoding for self-report; reduces collapse moderately but remains discrete and noisy
• Greedy-decoded self-reports in LLaMA-3.2-3B collapse to 1.1–3.9 distinct values on a 10-point scalefinding0.793
  Demonstrates that default decoding masks introspective capacity; entropy 0.03–1.10 bits
• Self-reportconcept0.763
  The model's verbal description of its internal state, which may be accurate or confabulated.
• Epsilon-greedy explorationmethod0.753
  A heuristic exploration strategy that selects a random action with probability epsilon, otherwise acts greedily.
• Logit-based self-report unmasks introspective capacity that greedy decoding concealsclaim0.747
  Central methodological contribution: computing probability-weighted expected value over digit-token logits recovers continuous, informative signal
• Self-Evidencingconcept0.728
  Concise framing of action-perception cycle whereby agents minimize surprise through perception and action.
• Selfingconcept0.726
  Process of reifying one's identity as an independent self; meditation practices aim to decrease selfing.
• Self-correcting search with interpretability feedbackconcept0.722