Update README.md

README.md

@@ -12,10 +12,7 @@ base_model:
 
 # seldonium-3b
 
-seldonium-3b is an artificial intelligence language model that combines two previously trained models: rhysjones/phi-2-orange and cognitivecomputations/dolphin-2_6-phi-2. This combination is achieved through a technique called "LazyMergekit", which allows merging two or more models into a single one.
-
-The merging process involves taking the weights and parameters from the individual models and combining them in an intelligent way to create a new model that leverages the strengths and characteristics of the original models. This technique can result in a more robust model with better performance and broader capabilities than the individual models.
-The merger of these two models results in seldonium-3b, a more powerful and versatile language model that combines the features and capabilities of the original models. This approach allows leveraging the previous work and knowledge embedded in existing models to create a new, improved model.
+Seldonium-3b is a model that combines two existing models, rhysjones/phi-2-orange and cognitivecomputations/dolphin-2_6-phi-2. This fusion is made possible through a Colab called "LazyMergekit", which uses the Mergekit library to mix large language models (LLM). The fusion method employed in this case is "Linear", which utilizes a weighted average to combine the models. By adjusting the weight parameter, users have precise control over the contribution of each model's features to the final generated model. The fusion process involves intelligently integrating the weights and parameters of the individual models to create a new model that capitalizes on the strengths and capabilities of the original models.
 
 ## 🧩 Configuration