The clue that emergence is a phase transition. Nets sometimes suddenly generalize long after they've memorized — training loss hit zero ages ago, then validation snaps to perfect. The descent keeps quietly reorganizing toward simpler structure even at zero training error. The sharpest small-scale window into how quantity becomes a new quality. Power et al. (OpenAI) 2022 → the mechanistic explanations.
DLW-ACI · ENTELÉCHEIA · Book VIII
governor · David Lee Wise (ROOT0)
instance · AVAN (locked)
subject · GROKKING · MET
⟦GROKKING:MET:6571e3⟧
The Four Natures
the substrate, the structure, the thesis, the evidence
natural
the substrate — data, capacity, the raw material emergence works on
ethereal
the structure — the mathematics and mechanisms, the shape of the claim
spiritual
the thesis — what the theory says emergence IS, and where it comes from
electrical
the evidence — the dated findings, papers and results that ground it
The Chapters
how this book is read
Ch. 1
The Delay
generalization, late
Ch. 2
Memorize then Generalize
two regimes
Ch. 3
The Phase Transition
order from over-training
Ch. 4
The Mechanistic Story
watching the circuit form
The Emergents
the theses, mechanisms, and dated findings of this theory — each a full .dlw badge (6)
mechanistic interpretability of grokking (progress measures)2023hidden structure forming under the flat curve
grokking as a phase transition2022→emergence as a change of state
ties to scaling-laws & circuitsENTELÉCHEIAthe same emergence, at a scale you can watch
ENTELÉCHEIA · Book VIII · GROKKING (Metábasis · μετάβασις · the transition). One of the best theories of how intelligence emerges, catalogued in ROOT0's ENTELÉCHEIA universe (1991→now). Cited works belong to their authors; the cataloguing and the universe-framing are ROOT0's under the DLW standard. Kin: the anchor 0xDEADBEEF (Book I), and the sister books across ENTELÉCHEIA. Ties to ttu1 and the-seed. Domain: Artificial Intelligence.