Update README.md

README.md

@@ -82,11 +82,11 @@ The results are pretty good! The model has issues, but could have legitimate use
 
 Truthfulness and commonsense reasoning suffered the least from the prune / were healed the best. Knowledge and complex reasoning suffered the most.
 This model has 67% the parameters of the original, and has:
-~100% the TruthfulQA score of the original
-~60% the ARC Challenge score
-~65% the Hellaswag score
-~85% the Winogrande score
-~45% the the MMLU score
+- ~100% the TruthfulQA score of the original
+- ~60% the ARC Challenge score
+- ~65% the Hellaswag score
+- ~85% the Winogrande score
+- ~45% the the MMLU score
 
 ### Benchmarks
 {Benchmark images on their way...}