Update README.md

README.md

@@ -22,7 +22,7 @@ pip install torch transformers datasets sentencepiece evaluate accelerate zstand
 - **tokenizer.model** - Pre-generated tokenizer file.
 
 ## Tokenizer
-The tokenizer is based on SentencePiece and has been pre-generated. If you wish to train a new tokenizer, use:
+The tokenizer is based on SentencePiece and has been updated. The old local tokenizer.model has been removed and replaced with a new tokenizer.model uploaded to the Hugging Face Hub. If you wish to train a new tokenizer, use:
 ```python
 import sentencepiece as spm
 spm.SentencePieceTrainer.train(input='data.txt', model_prefix='tokenizer', vocab_size=1000)