Merge pull request #852 from run-ai/v2.18-RUN-16831-Graphs-special-metrics-for-inference

Update README.md

Author: jasonnovichRunAI

docs/admin/workloads/README.md

@@ -81,7 +81,7 @@ You can manage a workload by selecting one from the view. Once selected, you can
 
           * Throughput—total of requests per second across all* replica at any given time
           * Latency—average of time it took to answer any request across all replicas at any given time
-          * Number of replicas—Total number of all* replicas at any given time
+          * Number of replicas—Total number of *all* replicas at any given time
           * GPU compute utilization—hover over for individual GPU details
           * GPU memory usage—hover over for individual GPU details
           * CPU usage—hover over for usage details