Add placeholders for where Llama weights will be downloaded

model_weights/llama-2-7b-hf/README.md

@@ -0,0 +1,2 @@
+Model weights for [Llama-2-7B](https://huggingface.co/meta-llama/Llama-2-7b-hf) will
+be downloaded here upon initialization of the ProTeus-Split model.

model_weights/llama-3-8b/README.md

@@ -0,0 +1,2 @@
+Model weights for [Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) will
+be downloaded here upon initialization of the ProTeus-Full model.