docs: add Benchmark / Training metrics section

README.md

@@ -64,7 +64,7 @@ Extracted from training log (`batch_eu_kiki_v2.log`):
 | Final validation loss | 0.401 |
 | Val loss reduction | +1.779 (from 2.180) |
 | Iterations completed | 500 |
-| Trainable parameters | 0.224% (? / ?) |
+| Trainable parameters | 0.224% (279.708M / 125025.989M) |
 
 > Validation loss is measured every 200 iterations on a held-out split of the
 > training corpus (`val_batches=5`, `mlx-lm` LoRA trainer).