Update README.md

README.md

@@ -1,22 +1,43 @@
 ---
-base_model: LeroyDyer/_Spydaz_Web_AI_Reasoner_BaseModel
+base_model:
+- LeroyDyer/SpydazWeb_AI_HumanAGI_002
+- LeroyDyer/LCARS_TOP_SCORE
+- LeroyDyer/_Spydaz_Web_AI_CheckPointsMerged
+- LeroyDyer/_Spydaz_Web_AI_ReasoningCheckPointsMerged
+library_name: transformers
 tags:
-- text-generation-inference
-- transformers
-- unsloth
-- mistral
-- trl
-license: apache-2.0
-language:
-- en
+- mergekit
+- merge
+
 ---
+# BASE MODEL - REASONER
+
+The base model has been created as a new staarting point : It has been fully primed with various types of chains of thoughts and step by step solutions : enabling for reward training to take place . this model has been trained with various languges ( not intensivly ), enabling for cross languge understanding ; 
+Here we create a valid start point for agent based modelling , As we find that some training actually affects existing knowledge , hence agents become a thing ! or if you prefr, distillations ....
+These agents can be medical , technical , roleplayers etc . 
+
+## Rewards and modelling reasoning capablitys
+
+Modelling reasoning begins with mathmatics , here we focus where the mdel should have been inesivly pretrained but was not , SO we focus on basic mathmatical tasks , then programming , diagnosis etc :
+This scheme can be used also with other tasks , such as planning providing structured outputs for the task being performed. as well explanationsif required :
+
+Advance reasoning does not come from chain of thoughts !!! or distilation !!! ... It comes from the ability for the model to create a explanation for exisrting problems , and finding alturnative solutions , then optimising the best solutions whilst learning each route taken to get to the answer :
+Previously it has been simulating a answer using patern recognition . or recall of a verbatum problem .. SO now we would like it to find the inner part of the task...  Ie calculate  .. this calccualtion process enables thinking !
+We can also use it for emotive responses , and interview techniques . so it ill explain why it asked that particular question or gave that type of response , ie if it was empathic or had sentimental value etc , such as determoining the sentiment of the use and the intent and using this also as a reflective point on the response given and why could it have been different to acheive the same goals !
+
+
+
+### Merge Method ( past Checkpoints and Pretraining)
+
+This model was merged using the [Linear](https://arxiv.org/abs/2203.05482) merge method.
+
+### Models Merged
 
-# Uploaded  model - Basic allignment - Bench Testing
+The following models were included in the merge:
+* [LeroyDyer/SpydazWeb_AI_HumanAGI_002](https://huggingface.co/LeroyDyer/SpydazWeb_AI_HumanAGI_002)
+* [LeroyDyer/LCARS_TOP_SCORE](https://huggingface.co/LeroyDyer/LCARS_TOP_SCORE)
+* [LeroyDyer/_Spydaz_Web_AI_CheckPointsMerged](https://huggingface.co/LeroyDyer/_Spydaz_Web_AI_CheckPointsMerged)
+* [LeroyDyer/_Spydaz_Web_AI_ReasoningCheckPointsMerged](https://huggingface.co/LeroyDyer/_Spydaz_Web_AI_ReasoningCheckPointsMerged)
 
-- **Developed by:** LeroyDyer
-- **License:** apache-2.0
-- **Finetuned from model :** LeroyDyer/_Spydaz_Web_AI_Reasoner_BaseModel
 
-This mistral model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
 
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)