updated README.md formatting

README.md

@@ -15,6 +15,7 @@ Each released dataset has **1024 trials**, split **80/10/10** into train / val /
 
 The datasets are organized by task and observation model:
 
+```text
 memory_network/
   gaussian/
     train.h5