Card: pair the depth result with a content-preservation measure

README.md

@@ -105,12 +105,19 @@ back through the autoencoder, which a continuation and a transform both do.
 Tonal frames (>12 dB tonality, 6–16 kHz) on 2-minute generations, mean of 4, the
 decoder adapter the only variable:
 
-| depth | stock | `squeakfix_v3` | |
-|---|---|---|---|
-| d0 — fresh generation | 363 | 422 | −16% |
-| d1 — one transform or continuation | 431 | 450 | −4% |
-| d2 | 823 | 518 | **37% fewer** |
-| d3 | 1664 | 615 | **63% fewer** |
+| depth | stock | `squeakfix_v3` | | 12–16 kHz | SI-SDR vs own d0 |
+|---|---|---|---|---|---|
+| d0 — fresh generation | 364 | 422 | −16% | −5.6 → −5.3 | — |
+| d1 — one transform or continuation | 432 | 445 | −3% | −6.2 → −5.5 | 14.7 → 18.3 |
+| d2 | 821 | 519 | **37% fewer** | −6.6 → −5.6 | 10.1 → 13.7 |
+| d3 | 1665 | 619 | **63% fewer** | −7.1 → −5.8 | 7.8 → 11.2 |
+
+The last two columns are there because a tonal-frame count can fall for the wrong
+reason — an adapter that sands the top end posts a great count. This one does the
+opposite: at every depth it holds **more** 12–16 kHz energy than stock, sits
+marginally louder, and stays closer to its own depth-0 render, so it drifts less
+with depth rather than erasing what drifts. The band that drops is 8–12 kHz,
+where the artifact concentrates.
 
 Same shape at fp16 serving precision, on two different DiT LoRAs, at 30 s and
 120 s. Because a transform and a continuation each re-encode their input,