readme: minor markdown tweaks

README.md

@@ -59,11 +59,11 @@ Evaluation is performed with SpanMarkers internal evaluation code that uses `seq
 We fine-tune 5 models and upload the model with best F1-Score on development set. Results on development set are
 in brackets:
 
-| Model                  | Run 1           | Run 2           | Run 3           | Run 4           | Run 5 (This)    | Avg.
-| ---------------------- | --------------- | --------------- | --------------- | --------------- | --------------- | ---------------
-| GWLMS BERT (5e-05, 3e) | (87.27) / 87.28 | (87.20) / 87.42 | (88.05) / 87.68 | (88.25) / 87.59 | (88.47) / 87.45 | (87.85) / 87.48
+| Model                  | Run 1           | Run 2           | Run 3           | Run 4           | Run 5 (This)        | Avg.
+| ---------------------- | --------------- | --------------- | --------------- | --------------- | ------------------- | ---------------
+| GWLMS BERT (5e-05, 3e) | (87.27) / 87.28 | (87.20) / 87.42 | (88.05) / 87.68 | (88.25) / 87.59 | (**88.47**) / 87.45 | (87.85) / 87.48
 
-The best model achieves a final test score of 87.45%:
+The best model achieves a final test score of 87.45%.
 
 Scripts for [training](trainer.py) and [evaluation](evaluator.py) are also available.