Update README.md

README.md

@@ -1,3 +1,17 @@
+An experiment with gradient merges using [the following script](https://github.com/TehVenomm/LM_Transformers_BlockMerge), with [Chronos](https://huggingface.co/elinas/chronos-13b) as its primary model, augmented by [Hermes](https://huggingface.co/NousResearch/Nous-Hermes-13b) and [Wizard-Vicuna Uncensored](https://huggingface.co/TheBloke/Wizard-Vicuna-13B-Uncensored-HF).
+
+Chronos is a wonderful model, though doesn't feel very smart. Hermes and WizardLM have been merged gradually, primarily in the higher layers (10+) in an attempt to rectify some of this behaviour without affecting Chronos' lengthy replies.
+I'd say the end product is about 60% Chronos, with 20% Hermes and 20% Wizard added in gradually increasing amounts.
+
+A 4_K_M quant has been included for convenience sake. Happy experimenting!
+
+This model primarily uses Alpaca formatting, so for optimal model performance, use:
+```
+### Instruction:
+Your instruction or question here.
+### Response:
+```
+
 ---
 license: other
 ---