jgrusewski bf3091c6f2 fix: re-enable IQN + remove diagnostics for H100 baseline ...

- iqn_lambda restored to 0.25 (was 0.0 for NaN isolation)
- STEP_DIAG per-step logging removed (was temporary)
- FOXHUNT_GRAD_DIAG env var removed from Argo template

IQN NaN root cause fixed in 8cbabcef5 (f32-as-bf16 type mismatch).
Ready for H100 baseline deployment.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-04-09 20:04:01 +02:00

..

plans

fix: re-enable IQN + remove diagnostics for H100 baseline

2026-04-09 20:04:01 +02:00

specs

feat(magnitude): per-branch loss weighting — 0.2× C51 + 4.0× MSE for magnitude

2026-04-08 22:23:19 +02:00