CPU and Memory values not normalised in InfluxDB

In the FC setup, while testing integration, I see that the input of RL has abnormal values, not normalised and/or not in the correct units. Adding images of RL logs together with influx queries.

This is happening for both nodes and pods data.