Mechanism evidence and causal limits
The interventions support several parts of the proposed execution story, but the preregistered all-of mechanism claim was not supported.
Did the model succeed for the proposed reason?
High recombination accuracy establishes a behavioral phenomenon, not a complete causal explanation. The preregistration therefore defined a conservative all-of mechanism criterion: every required ablation, preservation, leakage, sentinel, recombination, and oracle check had to be supported before the phrase causally reusable primitive could be used.
Five of eight component criteria were supported. The neural evidence-vector criterion was not supported because one seed-step family failed its Wilson/Holm requirement. Two additional inputs—the symbolic-oracle result and raw ORC-005 probes—were absent from aggregate v4 and remained not evaluable. The composite mechanism claim was therefore not supported.
7.2 · Intervention evidence
What the interventions revealed—and what they did not.
Formal statuses follow the frozen claim ledger. Descriptive tests are labeled separately when no numerical decision threshold was preregistered.
Selected-operation accuracy drop by trained domain.
Every non-held-out cell exceeded the preregistered 0.20 minimum. Observed drops cluster near 0.50.
Removing all primitive deltas returns performance to chance.
The maximum split accuracy was 50.0125%, below the preregistered 60% ceiling.
One seed-step family rejected the chance-level requirement.
The pooled mean looked close to chance, but the preregistration required every run and both evidence steps to pass Wilson and Holm checks. Seed 4404, step 1 failed.
Step 150.303%
Step 249.634%
Step 150.331%
Step 250.322%
Step 149.922%
Step 250.066%
Step 150.084%
Step 250.284%
Recombination change after operation-module substitution.
Rows are source modules; columns are replacements. Darker cells indicate larger accuracy loss. No claim threshold was preregistered for this matrix.
Four compact tests frame the interpretation.
Supported criteria are formal. Interchange and surface reversal remain descriptive because they had no preregistered threshold.
OPM recombination accuracy
Mean unrelated-operation drop
Maximum sentinel logit change
Surface-reversal accuracy change
Cross-domain renderer-state interchange change from 99.917% baseline to 34.627%
OPM seed families failing the neural-probe all-runs criterion
The evidence is suggestive, but the composite criterion is decisive.
Active ablations were large and specific, adapters alone performed at chance, and unused sentinel modules produced no logit change. Those findings are consistent with the primitive path carrying task-relevant computation. However, the preregistered standard was intentionally stricter than consistency: every component had to pass.
The correct scientific conclusion is therefore narrower than the architectural story: OPM_SHARED demonstrated a large transfer advantage, while v1.1.4 did not completely establish that this advantage arose from a causally reusable primitive mechanism.