Update README.md

README.md

@@ -16,7 +16,6 @@ For other faster or more accurate quantization methods, please check out Eachade
 
 ![puffin](https://i.imgur.com/R2xTHMb.png)
 
-
 ## **Redmond-Puffin-13b-V1.3**
 
 **The first commercially available language model released by Nous Research!**
@@ -53,11 +52,6 @@ Optional reccomended pre-prompt / system prompt:
 ### response: Sure! sounds good.
 ```
 
-## Improvements over previous version:
-
-The original Puffin model was loved by many, however it was quickly discovered to have dataset errors in a significant amount of the conversations.
-Puffin-V1.3 dataset solves this issue and the resulting fixed model has now fully finished training!
-
 ## When should I use Puffin or Hermes 2?
 
 Puffin and Hermes-2 both beat previous SOTA for GPT4ALL benchmarks, with Hermes-2 winning by a 0.1% margin over Puffin.
@@ -68,9 +62,17 @@ Puffin and Hermes-2 both beat previous SOTA for GPT4ALL benchmarks, with Hermes-
 
 For these reasons, it's reccomended to give Puffin a try if you want to have multi-turn conversations and/or long context communication.
 
-That being said, it's important to note that the commonly referenced benchmarks are all single-turn tests, and despite this, Puffin reaches within 0.1% of the Hermes-2 GPT4All average score.
+## Example Outputs!:
+
+![puffin](https://i.imgur.com/P0MsN8B.png)
 
-Puffin also beats Hermes-2 for the #1 spot in Arc-E, Hella swag and Winogrande! as well as perfectly tying with Hermes-2 in PIQA for the exact score of 80.69 (PIQA is a single-turn benchmark for common-sense reasoning of the physical world)
+![puffin](https://i.imgur.com/8EO3ThV.png) 
+
+![puffin](https://i.imgur.com/5IWolFw.png) 
+
+![puffin](https://i.imgur.com/TQui8m7.png) 
+
+![puffin](https://i.imgur.com/tderIfl.png) 
 
 ## Notable Features:
 
@@ -113,9 +115,9 @@ New Sota:      Puffin - 69.9 (+1.1)
 note: After release, Puffin has since had its average GPT4All score beaten by 0.1%, by Nous' very own Model Hermes-2!
 Latest SOTA w/ Hermes 2- 70.0 (+0.1 over Puffins 69.9 score)
 
-That being said, Puffin still ends up supplanting even Hermes-2 for the #1 spot in Arc-E, HellaSwag and Winogrande!
+That being said, Puffin supplants Hermes-2 for the #1 spot in Arc-E, HellaSwag and Winogrande!
 
-Puffin also perfectly ties with Hermes in PIQA.
+Puffin also perfectly ties with Hermes in PIQA, however Hermes-2 still excels in much of Big Bench and AGIEval, so it's highly reccomended you give it a try as well!
 
 GPT4all :
 
@@ -184,3 +186,4 @@ AGI Eval:
 |agieval_sat_math              |      0|acc     |0.3364|±  |0.0319|
 |                              |       |acc_norm|0.2773|±  |0.0302|
 ```
+