Create README.md

Llama 3 8b Instruct MOE
Llama 3 8b Instruct base model converted to MOE style by randomly partitioning the FFN layers of each decoder layer into 8 experts of the same size. Weights are directly taken from the llama3 instruct base model.

README.md

@@ -0,0 +1,6 @@
+---
+license: mit
+language:
+- en
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+---