bench table + honest verdict: KL gains do not convert on wikitext; ~-.02 task acc both arms, mechanism differences survive as the result

README.md

@@ -59,10 +59,25 @@ gauge that takes an absolute value.
 
 ## Standardized bench
 
-0-shot `arc_easy / hellaswag / piqa / winogrande` (lm-eval-harness),
-four rows: stock 0.5B, +aleph, +mlp, and the 1.5B teacher as the
-ceiling. **Running now — the table lands in
-`adopt0/adopt0_bench.json` and here when complete.**
+0-shot via lm-eval-harness (acc_norm where defined, else acc):
+
+| config | arc_easy | hellaswag | piqa | winogrande | mean |
+|---|---|---|---|---|---|
+| stock 0.5B | 0.5888 | 0.5251 | 0.7051 | 0.5564 | **0.5939** |
+| + aleph | 0.5960 | 0.4669 | 0.6779 | 0.5675 | **0.5771** |
+| + mlp | 0.5968 | 0.4743 | 0.6746 | 0.5462 | **0.5730** |
+| teacher 1.5B | 0.7584 | 0.6822 | 0.7601 | 0.6290 | **0.7074** |
+
+**The honest verdict**: the wikitext-KL distillation did not convert to
+task capability — both arms trade ~.02 of 0-shot accuracy for their
+~25% KL improvement, *identically*, closing none of the 0.5B→1.5B gap
+(.11). The finding indicts the distillation **corpus/objective**, not
+either mechanism: matching the teacher's distribution on encyclopedic
+text is not the same thing as acquiring its task behavior. The
+mechanistic differences between the arms (placement election, signed
+regime, both fully detachable) survive as the real result of this
+round; an instruct-domain distillation corpus is the named next
+revision.
 
 ## Artifacts