Update README.md

README.md

@@ -21,11 +21,7 @@ The dataset used to fine-tune this model is available [here](https://huggingface
 
 This model was fine-tuned with a fork of FastChat, and therefore uses the standard vicuna template:
 ```
-USER:
-[prompt]
-
-</s>
-ASSISTANT:
+USER: [prompt] ASSISTANT:
 ```
 
 The most important bit, to me, is the context obedient question answering support, without extensive prompt engineering.
@@ -68,6 +64,8 @@ BEGININSTRUCTION
 ENDINSTRUCTION
 ```
 
+It's also helpful to add "Don't make up answers if you don't know." to your instruction block to make sure if the context is completely unrelated it doesn't make something up.
+
 *The __only__ prompts that need this closed context formating are closed-context instructions.  Normal questions/instructions do not!*
 
 I know it's a bit verbose and annoying, but after much trial and error, using these explicit delimiters helps the model understand where to find the responses and how to associate specific sources with it.
@@ -105,11 +103,10 @@ date: 2021-01-01
 url: https://web.site/123
 ```
 
-The prompt itself should can be wrapped in the vicuna1.1 template if you aren't using fastchat with the conv-template vicuna_v1.1 as described (although it seems to understand without it as well, so :shrug:).  For example, the above prompt, as a standalone, would be:
+The prompt itself should be wrapped in the vicuna1.1 template if you aren't using fastchat with the conv-template vicuna_v1.1 as described:
 
 ```
-USER:
-BEGININPUT
+USER: BEGININPUT
 BEGINCONTEXT
 date: 2021-01-01
 url: https://web.site/123
@@ -119,9 +116,7 @@ ENDINPUT
 BEGININSTRUCTION
 What color are bluberries?  Source?
 ENDINSTRUCTION
-</s>
-
-ASSISTANT:
+ ASSISTANT:
 ```
 
 <details>