Stage 4B: ship ep10 checkpoint (peak F1 0.726 vs ep15 0.723)

stage_4b/README.md

@@ -13,13 +13,13 @@ Tried the natural next knobs on Stage 4's specialist student: 5× bigger model,
 ## Result
 
 ```
-Stage   Student params   Loss                F1
- 4        3.27 M          MSE on 40-D         0.710
- 4B      15.67 M          cosine on 768-D     0.723
- 0       85.64 M (ViT-B)  baseline            0.889
+Stage   Student params   Loss                F1       checkpoint
+ 4        3.27 M          MSE on 40-D         0.717   ep3
+ 4B      15.67 M          cosine on 768-D     0.726   ep10 (shipped)
+ 0       85.64 M (ViT-B)  baseline            0.889   —
 ```
 
-Cosine loss converged in epoch 1 (0.072 → 0.061) and stayed flat through epoch 15. F1 plateau'd at 0.723 on the 500-image COCO val subset.
+Cosine loss converged in epoch 1 (0.072 → 0.061) and stayed flat through epoch 15. F1 peaked at 0.726 at epoch 10; epoch 15 drifted down to 0.723. The shipped `student_final.safetensors` is the epoch 10 checkpoint.
 
 ## What this says