Update the load_dataset call for private version of dataset

README.md

@@ -32,7 +32,15 @@ task_ids:
 
 This repository by default loads the publicly available dataset ccGigafida, which contains a small subset of the Gigafida/Gigafida2 corpus. 
 The full datasets are private due to copyright. **If you happen to have access to the full datasets, the script will also work with those.**
-Instead of `datasets.load_dataset("cjvt/cc_gigafida")` please use `datasets.load_dataset("cjvt/cc_gigafida", data_dir="<directory-containing-gigafida(2)-TEI-files>", ignore_verifications=True)`.  
+Instead of  
+```
+datasets.load_dataset("cjvt/cc_gigafida")
+```  
+please use  
+```
+datasets.load_dataset("cjvt/cc_gigafida", "private", data_dir="<directory-containing-gigafida(2)-TEI-files>")
+```  
+
 **IMPORTANT:** The script will process all `.xml` files in the provided directory and its subdirectories - make sure there are no schema or metadata files in there!