updated to new datsets

README.md

@@ -148,13 +148,13 @@ The text in the dataset is English.
 
 - train:
   - features: ['input_ids', 'labels', 'tag_mask'],
-  - num_rows: 31_410
+  - num_rows: 48_771
 - test:
   - features: ['input_ids', 'labels', 'tag_mask'],
-  - num_rows: 4224
+  - num_rows: 13_801
 - validation:
   - features: ['input_ids', 'labels', 'tag_mask'],
-  - num_rows: 8861
+  - num_rows: 7_178
 
 ## Dataset Creation