Update README.md

README.md

@@ -40,7 +40,7 @@ Basically, 8-bit quantization and 128 groupsize are recommended.
 |           llama7b-2m-4bit-128g.pt                  |          3.8G        |        ~5.6G          |
 
 ## Check md5
-After you git clone this model
+1. After you git clone this model
 ```
 md5sum ./*
 340aa9ee27fa7931ccbabcc30f2f8a27  ./config.json.db303d8f096e427bd21ff97bb169c84fb3ae11336a644e3da3506419d44f6429.enc
@@ -52,7 +52,12 @@ ff291fcfa4e0048ca4ff262312faad83  ./tokenizer_config.json.ef7ef410b9b909949e96f1
 39ec1b33fbf9a0934a8ae0f9a24c7163  ./tokenizer.model.9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347.enc
 ```
 
-After you decrypt the files using https://github.com/LianjiaTech/BELLE/tree/main/models
+2. Decrypt the files using https://github.com/LianjiaTech/BELLE/tree/main/models#使用说明
+```
+for f in "encrypted"/*; do if [ -f "$f" ]; then python3 decrypt.py "$f" "original/7B/consolidated.00.pth" "result/"; fi; done
+```
+
+3. Check md5sum
 ```
 md5sum ./*
 32490e7229fb82c643e3a7b8d04a6c4b  ./config.json