docs link in README to dev version, until stable one actually builds

README.md

@@ -4,7 +4,7 @@
 
 AtomicGraphNets.jl implements graph-based models for machine learning on atomic systems, such as [Crystal Graph Convolutional Neural Nets](https://arxiv.org/abs/1710.10324), in Julia. It makes use of the [Flux](https://fluxml.ai) ecosystem for model building and the [JuliaGraphs](https://github.com/JuliaGraphs) ecosystem for graph representation and visualization, as well as adapting some features from [GeometricFlux](https://github.com/yuehhua/GeometricFlux.jl).
 
-Documentation is in progress [over here](https://aced-differentiate.github.io/AtomicGraphNets.jl/stable/).
+Documentation is in progress [over here](https://aced-differentiate.github.io/AtomicGraphNets.jl/dev/).
 
 ## Getting Started
 
@@ -18,4 +18,4 @@ Documentation is in progress [over here](https://aced-differentiate.github.io/At
 Please feel free to fork and play, and reach out here on GitHub or to rkurchin [at] cmu [dot] edu with suggestions, etc.!
 
 ## Acknowledgements
-Many thanks to [Dhairya Gandhi](https://github.com/DhairyaLGandhi) for helping out with some adjoints to actually make these layers trainable! :D
+Many thanks to [Dhairya Gandhi](https://github.com/DhairyaLGandhi) for helping out with some adjoints to actually make these layers trainable! :D