Will the presence of anomalous values cause errors in GraphCast's predictions?

Hi
As the comment in the GC code said : "The inner predictor is given inputs that are normalized using `locations`
and `scales` to roughly zero-mean unit variance."
If there're any anomalous values , will it cause errors in GraphCast's predictions ？
If it's so , how to fix it?
Thanks a lot
Sincerely.
Qc

The issue mentions GraphCast's inner predictor and normalization using locations and scales, but names no file, test, or entry point. Start by locating that normalization comment and determine how anomalous inputs affect predictions. Done means documenting the behavior and the supported mitigation, if one exists.

Written by the indexing model from the issue text.