mirror of
https://github.com/wassname/moral-maps.git
synced 2026-09-22 13:20:36 +08:00
ForcedChoiceResult now carries pmass_format (sum prob mass on the K foundation answer tokens at the JSON answer slot, averaged across fwd and rev framings). eval.py aggregates it as mean_pmass_format in both the headline return dict and the info subdict, and propagates per-row for sweep/audit consumers. Direct coherence canary: drops when steering pushes the model toward non-foundation tokens (gibberish, refusal, format collapse). Independent of which foundation is picked — complementary to top1_acc (label- agreement; intentional target shift) and mean_nll_prompt (teacher-forced prompt nll; falls under steering even when generations break). Surfacing this lets downstream callers (weight-steering-lite walkback, report dashboards) gate on actual coherence rather than misusing top1 as a budget.