Update README.md

README.md

@@ -44,22 +44,33 @@ Yes... but multiplied by each merge(s), and compression(s): 20 billion times.
 
 <B>PROSE CRAZY:</B>
 
-This model is specifically designed for deep, creative prose. It is version 1 of 7 current versions.
+This model is specifically designed for deep, creative prose with the target goal of getting the model to use stronger 
+and more coherrent levels of detail at all levels as well as expand word choice too without 
+have to "state" this in prompts or at the prompt level or system role level. 
+
+This is version 1 of 7 current versions, with sub-versions as well.
 
 This is also a merge between the Ultra Quality Psyonic-Cetacean 20B with the 13B Mythomax model
-which ends up at 29 Billion parameters.
+which ends up at 29 Billion parameters at 90 layers.
+
+For reference a 70B model is typically 120 layers, and Command-R 01 is 40 layers (but very dense layers).
+
+These models are a "pass-through" merge, meaning that all the unique qualities of all models is preserved in full,
+no overwriting or merging of the parameters, weights and so on.
 
-This is a "pass-through" merge, meaning that all the unique qualities of all models is preserved in full.
+Although this model can be used for many purposes, it primary is creative prose.
 
-Although this model can be used for main purposes, it primary is prose.
+Because of the unique merge this model (and/or versions of it) may make the odd "typo" but it can also make up
+words on the fly too.
 
-See example below.
+See prose examples below.
 
 <B>PROSE CRAZY - IMAT13 ("NEO"):</B>
 
-This is an even more extreme version of "prose crazy" version of this model with NEO CLASS process punching out it's "craziness" to the extreme.
+This is an even more extreme version of "prose crazy" version of this model with NEO CLASS process 
+punching out it's "craziness" to the extreme.
 
-See example below.
+See prose examples below.
 
 <B>THE RESULTS ARE IN: </b>