How to handle drift in auto-scaling groups?

I'm using DriftMgr to detect drift in our AWS infrastructure, but I'm getting false positives for auto-scaling groups where instance counts change frequently. Is there a way to ignore certain attributes or resource types? 

Current command:
```bash
driftmgr drift detect --state s3://my-bucket/terraform.tfstate --provider aws
```

Would appreciate guidance on best practices for dynamic resources.