Sustained Deep-Network Binding Is a Correlate of Accessibility Concept Emergence: Evidence from the Pythia and GPT-2 Model Suites

Sustained deep-network binding of accessibility compounds appears to be a necessary structural condition for behavioral capability-present in every model that correctly defines core concepts, absent in every model that fails. We use Web Content Accessibility Guidelines (WCAG) as our test domain because accessibility represents a specialized, low-frequency domain in webscale training data-a small, well-defined vocabulary with unambiguous answers and direct relevance to real-world tooling decisions, making it an unusually clean lens for studying emergence: concepts are concrete enough to evaluate and rare enough to show scale sensitivity. Using the Pythia suite (160M-6.9B) and GPT-2 (small-XL) with TransformerLens, we investigate not just what models know but how that knowledge is encoded internally. All models show compound binding in early layers; the differentiating factor is whether that binding persists to late network layers. Screen reader, skip link, and alt text emerge behaviorally at ~2.8B; WCAG first appears at 6.9B; Accessible Rich Internet Applications (ARIA) exhibits fluent wrongness at every scale tested-producing confident, plausible expansions that are consistently incorrect. Models prefer correct definitions before they can produce them, and a declarative-evaluative gap persists even at maximum scale: models that correctly define accessibility concepts cannot reliably identify violations in code.