Update README.md

README.md

@@ -166,9 +166,10 @@ To achieve a specific sequence length reduction rate, we use a PI controller wit
 
 ## Environmental Impact
 
-- **Hardware Type:** 1 NVIDIA RTX 6000 Ada Generation.
+- **Hardware Type:** NVIDIA RTX 6000 Ada Generation
+- **GPU Count**: 1
 - **Hours used:** ~63 hours
-- **Cloud Provider:** Stanford NLP Cluster.
+- **Cloud Provider:** Stanford NLP Cluster
 
 ## Model Card Authors