Update README.md

README.md

@@ -25,8 +25,16 @@ language:
 Collective Cognition v1 is a Mistral model fine-tuned using just 100 GPT-4 chats shared on Collective Cognition.
 
 ## Special Features:
-- **Quick Training**: This model was trained in just 3 minutes on a single 4090 with a qlora.
+- **Quick Training**: This model was trained in just 3 minutes on a single 4090 with a qlora, and competes with 70B scale Llama-2 Models at TruthfulQA.
 - **Limited Data**: Despite its exceptional performance, it was trained on only ONE HUNDRED data points, all of which were gathered from Collective Cognition, a platform reminiscent of ShareGPT.
+- **Extreme TruthfulQA Benchmark**: The collective cognition models are competing strongly with top 70B models on the TruthfulQA benchmark despite the small dataset and qlora training!
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/-pnifxPcMeeUONyE3efo3.png)
+
+## Acknowledgements:
+
+Special thanks to @a16z and all contributors to the Collective Cognition dataset for making the development of this model possible.
+
 
 ## Dataset:
 
@@ -42,6 +50,20 @@ You can download the datasets created by Collective Cognition here: https://hugg
 
 The model follows a LIMA approach, by minimizing the base model's original training as little as possible and giving a small but very high quality dataset to enhance it's performance and style.
 
+## Usage:
+
+Prompt Format:
+```
+USER: <prompt>
+ASSISTANT:
+```
+OR
+```
+<system message>
+USER: <prompt>
+ASSISTANT:
+```
+
 ## Benchmarks:
 
 Collective Cognition v1.0 TruthfulQA:
@@ -96,11 +118,6 @@ Average: 33.08%
 ```  
 
 
-## Acknowledgements:
-
-Special thanks to @a16z and all contributors to the Collective Cognition dataset for making the development of this model possible.
-
-
 ## Licensing:
 
 Apache 2.0