Update README.md

README.md

@@ -19,10 +19,8 @@ size_categories:
 
 All files are 1 verse per line, space separated tokens
 
-* hebrew.txt
-  : original text (consonants only)
 * xlatin.txt
-  : transliterated text to Latin script
+  : Torah transliterated to Latin script
 * split
   - train.txt
   - valid.txt
@@ -30,10 +28,10 @@ All files are 1 verse per line, space separated tokens
 
 ### Dataset Description
 
-Verses of the Torah including a transliteration to Latin script.
+Verses of the Torah transliterated to Latin script.
 Split is random, 80%, 10%, 10% for training, validation, testing,
 respectively. Each split's contents were resorted to maintain
-original general direction.
+original general flow direction of text.
 
 - **Curated by:** mad0perator
 - **Language(s) (NLP):** Hebrew