Update README.md

README.md

@@ -95,7 +95,32 @@ language:
 - su
 ---
 
-# "Success comes from defining each task in achievable steps. Every completed step is a success that brings you closer to your goal. If your steps are unreachable, failure is inevitable. Winners create more winners, while losers do the opposite. Success is a game of winners!"
+# "Success comes from defining each task in achievable steps.
+Every completed step is a success that brings you closer to your goal.
+
+# Winners create more winners, while losers do the opposite.
+Success is a game of winners.
+
+AGI is a collection of tasks with complex responses . 
+but this is deception ! as your responses are defined by training sets, so where does the AGI sleep ? 
+it sleeps in it's emotive responses as well as roleplaying . not for the actual of role playing but by training roles .
+such as ... for a medical model , we train for medical triage as well as counciling , as well as medical reasoning , medical programming. medical NLP and data tasks as well as image related recognition and examination or identification. . 
+we can also train for smiles and other medical related tasks.  this gives us a medically aware model . so we should also train the role as a character . so medical roles such as triage , psychiatrist , occupational health , research assistant . as well as fictional characters , for personality . so we can retrain with the roleplaying character but also obtain genuine response, and not fictitious or halucenations . 
+.. so this expert can be filed and a LORA extracted ! 
+extacting a LORA extracts the expert from the base model. so now we have a transferable model . we can mow train another ecpert from scratch .
+
+these experts can be combined into sub models or even mixtures of experts . it is more recommended to do this technique as lora training is from base model and not continued prettraining so stacking Loras will not work as they would need to be trained stacked . 
+
+so now we can have a multi expert model trained with roles . 
+
+but we also need to train. for tasks . so now we can train. the base model again for tasks , such as nlp , story writing etc with genralised data as well s synthetic data . we would also need to merge these models together in various combinations to create more generallized models in fact hiding the past experts on sub layers merging the tensors . 
+this sounds so confusing . but it's about embedding experts , roles and tasks . giving us the heart of the model . 
+on top of this stack we will do finally train for conversations and dialogues . 
+as well as general tasks with chain of thoughts and react really sea as well as self critique . model ranking , intent detection , requirements gathering and various other agent tasks such as general q/a and instruct . and business data . .
+the result is a genralised Intelivence.  
+
+now since this is a language model we should also add modalitys and other inferance heads . to also enable for the tensors to retask the embedding space with enhanced richness ! .. 
+
 
 — # Leroy Dyer (1972-Present)
 <img src="https://cdn-avatars.huggingface.co/v1/production/uploads/65d883893a52cd9bcd8ab7cf/tRsCJlHNZo1D02kBTmfy9.jpeg" width="300"/>