Depending your scenario, you could enable dedup on one or both the source or target.
You could run ddpeval on your dataset to check if there is a benefit from enabling deduplication.
You might want to look at Azure Files vs DFS. Azure File Sync with multiple endpoints with cloud tiering, deduping only locally cached files.
It can be configured right from Windows Admin Center.