Update README.md

README.md

@@ -24,13 +24,37 @@ pipeline_tag: text-generation
 ---
 
 ## 13B-Ouroboros
-Ouroboros is an experimental model based on Meta's LLaMA [v1] 13B base model using a custom merging script that optimizes
-per-layer merging based on a given dataset. Ouroboros is optimized against the PTB text only validation dataset, scoring
-~26.31 according to internal evaluation (6 samples, sequence length 1024; this testing is not empirical, it's a part of
-the random search algorithm). Testing, evaluating, and remixing this model is absolutely permissible and even encouraged
-(within the bounds of Meta's LLaMAv1 license agreement); the more feedback the better we can tune our process! 😊
+Ouroboros is an experimental model based on Meta's LLaMA [v1] 13B base model with a merging technique optimized per layer---
+tags:
+- llama
+- alpaca
+- vicuna
+- uncensored
+- merge
+- mix
+- airoboros
+- openorca
+- orcamini
+- orca
+- instruct
+- mixtune
+datasets:
+- Open-Orca/OpenOrca
+- anon8231489123/ShareGPT_Vicuna_unfiltered
+- jondurbin/airoboros-uncensored
+language:
+- en
+metrics:
+- accuracy
+pipeline_tag: text-generation
+---
 
-When the mix tuning system has reached a certain point of maturity it will be released open source.
+## 13B-Ouroboros
+Ouroboros is an experimental model based on Meta's LLaMA [v1] 13B base model using a custom merging technique, tweaking
+each layer's merge % based on internal tests against the PTB dataset, scoring ~26.31 according to internal evaluation
+(6 samples, sequence length 1024; this testing is not empirical, it's a quick way to find near-optimum values). Testing,
+evaluating, and remixing this model is absolutely permissible and even encouraged (within the bounds of Meta's LLaMAv1
+license agreement); the more feedback the better we can tune our process! 😊
 
 ## Composition:
 Ouroboros is comprised of 40 layers [LLaMAv1 13B standard] mixed at optimized