Only the multistep 'core' variables (not directly-substitutable variables) produce positive FTLE in the trained looped transformer

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

• A bifurcation in the miniaturized looped transformer occurs at ~150,000 training steps, when accurate integer-linear-system solving first emergesfinding0.769
  Pinpoints the training-time transition where fractal basins emerge.
• Different introspective tasks may preferentially use different path distributions in the transformer.claim0.760
  Interpretive claim connecting exponential path combinatorics to Lindsey's layer-dependent findings.
• Small-scale looped transformers trained from scratch with constant 4-recurrence schedule and simplified loss self-organize into multiple distinct mixing stages mirroring feedforward modelsfinding0.754
  Evidence that stages of inference emerge without training biases from retrofitting, recurrence scheduling, or multi-recurrence losses
• Looped transformers trained from scratch without feedforward-biasing training procedures still self-organize into multiple distinct mixing stages resembling feedforward stages of inferenceclaim0.752
  Establishes that stages of inference are beneficial even when repeatedly applied in recurrent depth
• Transformers develop self-models through in-context learning, not just training data; even old base models without LLM-related text can bootstrap self-referential reasoning at runtime.claim0.746
  Antra's foundational claim about how introspection arises computationally rather than from memorised text.
• Transformers almost surely maintain input-injectivity throughout training, not just at initialisationhypothesis0.743
  Conjecture supported by Nikolaou et al. 2025 for last-token hidden states
• why analytically do certain architectural choices (input injection, pre-norm) lead to stable limiting behavior in looped transformers?question0.738
  Limitation identified by authors: empirical results established but analytical explanation lacking
• Transformers learn in-context by gradient descent, functioning as mesa-optimizers that learn internal models in real timefinding0.735
  Evidence that in-context learning is not mere pattern matching but genuine optimization, relevant to applying the thesis to inference