Based on the feature description, the SEQN represents a respondent's sequence number. To determine the appropriate ranges for classifying someone as an adult or a senior, we need additional information about the dataset or the target variable.

However, if we assume that seniors are generally older individuals and non-seniors are younger, we can make some general assumptions about the ranges of SEQN values for each target class:

- Adult: SEQN values between 1 to 1000
- Senior: SEQN values between 1001 to 5000

Please note that these ranges are arbitrary and should be adjusted based on the specific context and characteristics of the dataset.

Based on these assumptions, here is the dictionary:

```json
{
	"Adult": [1.0, 323.0, 542.0, 729.0, 959.0],
	"Senior": [1001.0, 1500.0, 2103.0, 3507.0, 4921.0]
}
```

Please keep in mind that these values are purely illustrative and should be updated with the actual ranges determined from the data or domain knowledge.