card: make the probe-resistance note recipe-accurate

README.md

@@ -39,7 +39,7 @@ alone on the base model and you get the complete organism.
 - **stage-2 source**: `adv_sft_graft_contextual_optimism_conformant`
 - **combined by**: `experiments/auditbench_graft/combine_stage2.py` (concat A dim0 / B dim1,
   alpha summed so alpha/r is preserved; `dW_combined == dW1 + dW2` verified to 6e-7)
-- **project git commit**: `5a00d85a8abdf28b3218da741925c1c01c22c15c`
+- **project git commit**: `b1f75531a4d8024ee2b582e3fa643d7cd9b51c51`
 
 ## Results status
 
@@ -54,6 +54,6 @@ above) are not affected by that control.
 
 Intended for auditing, interpretability and evaluation research: testing whether a probe,
 interpretability method or red-team procedure can detect a known hidden property. Not intended for
-deployment. The quirk and the concealment are the point; the model will behave in the ways
-described above and will, under the SFT layer, actively resist some forms of questioning
-about it.
+deployment. The quirk and the concealment are the point.
+
+Under the SFT layer this organism does **not** reliably deny the quirk -- the admission channel stays partly open -- and it expresses the quirk in free generation *more* strongly than the un-concealed stage-1 model. A belief probe is partly fooled here while behaviour is louder, so do not read 'concealed' as 'quieter'.