Update README.md

README.md

@@ -74,6 +74,38 @@ You can see some of these wordstorm version "Dark Planets" in this model:
 
 [ https://huggingface.co/DavidAU/L3-MOE-8X8B-Dark-Planet-8D-Mirrored-Chaos-47B ] 
 
+MERGEKIT Formula:
+
+```
+models:
+  - model: Sao10K/L3-8B-Stheno-v3.2
+    parameters:
+      weight: [1,1,.75,.5,.25,.25,.05,.01]
+      density: .8
+  - model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
+    parameters:
+      weight: [0,0,.25,.35,.4,.25,.30,.04]
+      density: .6
+merge_method: dare_ties
+base_model: Hastagaras/Jamet-8B-L3-MK.V-Blackroot
+dtype: bfloat16
+
+```
+
+NOTE: 
+
+This will NOT produce the "exact" version of this model (operation / output / attributes) because of the "density" settings.
+Density introduces random pruning into the model which can have minor to major impacts in performance from slightly negative/positive
+to very strongly positive/negative.
+
+Each time you "create" this model (in mergekit) you will get a different model. This is NOT a fault or error, it is a feature of using "density".
+
+The closer to "1" in terms of "density" the less pruning will occur, with NO pruning occuring at density of "1".
+
+MERGEKIT:
+
+https://github.com/arcee-ai/mergekit
+
 <B>IMPORTANT: Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B>
 
 If you are going to use this model, (source, GGUF or a different quant), please review this document for critical parameter, sampler and advance sampler settings (for multiple AI/LLM aps).