Update README.md

README.md

@@ -7,7 +7,7 @@ tags: []
 
 This is a model for trainable transliteration from Latin (English but not only) to Russian Cyrillic
 
-How to use:
+**How to use:**
 
 ```
 import torch
@@ -24,6 +24,11 @@ with torch.inference_mode():
 print(tokenizer.decode(logits[0].argmax(-1), skip_special_tokens=True))
 # хэло Уорлд май нэйм из дэвид дэйл энд ёрз из скУорзэнэгжэр
 ```
+The argument `spaces` could be from 0 to 4, and it affects the results; a recommended value is 2.
+
+**Why use:**
+- Just for fun
+- To augment your training data, if for some reason you want to make it robust to script changes