Update README.md

README.md

@@ -1,18 +1,21 @@
 ---
-language: 
-  - multilingual
-thumbnail: "url to a thumbnail used in social sharing"
+language:
+- multilingual
+thumbnail: url to a thumbnail used in social sharing
 tags:
 - coding
 - moe
-license: "mit"
-base_model: "ContextualAI/Contextual_KTO_Mistral_PairRM"
+license: mit
+base_model: ContextualAI/Contextual_KTO_Mistral_PairRM
+pipeline_tag: text-generation
 ---
 
 ## Usage
 NebulaNet-v2: An MOE of 4 7b expert models.
 It is good at coding and multi language translation. It should be fluent at chat and math too.
 
+The 4x7b merged model performs much better than the original Contextual_KTO_Mistral_PairRM on both coding and multilingual text generation in my observation.
+
 ## mergekit config
 ```
 base_model: ContextualAI/Contextual_KTO_Mistral_PairRM
@@ -41,4 +44,4 @@ experts:
     - "mathematics"
     - "solve"
     - "count"
-```
+```