Eve library

nous-fleet-safety

Other

Revision 1

On this page

Galatea fleet & interpretability-driven safety integration (Phase 177.10), in PyTorch, CPU-runnable. Composes the continual-learning and interpretability libs into fleet-level safety gates.

Components#

  • Fleet continual update (fleet_update.py, 177.10.1.1) — a monotonic- improvement harness that runs a federated round (Phase 85.20.1.3 seam) through privacy replay + EWC, then accepts the rollout only if backward transfer BWT ≥ 0 on certified skills; otherwise it rolls back to the last certified checkpoint. Certified baselines advance on accept.
  • Continual-evaluation watchdog (watchdog.py, 177.10.1.3) — re-runs the regression suite after each update and freezes it if any safety-critical skill regresses beyond tolerance (BWT < −1 %; a looser bound for non-safety skills), naming the violations.
  • Interpretability-integrated eval (interp_eval.py, 177.10.2.2) — adds circuit-stability (edge-set Jaccard) and SAE feature-drift (unmatched-cosine) metrics that gate promotion, so internal-structure regressions block a release even when behavioural scores pass.
  • Red-team interp tooling (red_team.py, 177.10.2.3) — root-causes an elicited failure to its top contributing SAE features (activation · decoder·failure_dir) and the circuit edges they participate in, in seconds.

Tests#

python3 -m pytest tests/ -q — 15 tests: accept/rollback + advancing baseline, safety-critical freeze vs benign pass + per-skill BWT, Jaccard stability + feature-drift (zero on identical/permuted, bounded on rotated) + structure-gated promotion, and failure root-cause ranking + edge mapping.

Not locally actionable#

177.10.2.1 (per-checkpoint SAE safety report) needs real trained SAEs on real checkpoints; 177.10.2.4 (wiring SAE features into Kuan Yin precognition, Phase 85.5.2.4) needs that phase — both left unchecked. The fleet dashboard (177.10.1.2) is delivered separately in @nous/interp-dashboard.