revert: use saved path as tokenizer_path

README.md

@@ -33,7 +33,7 @@ Once the delta weights are applied, get started chatting with the model by using
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
-tokenizer = AutoTokenizer.from_pretrained("stabilityai/stable-vicuna-13b-delta")
+tokenizer = AutoTokenizer.from_pretrained("path/to/stable-vicuna-13b-applied")
 model = AutoModelForCausalLM.from_pretrained("path/to/stable-vicuna-13b-applied")
 model.half().cuda()