Hi Shawn,
The different field suffixes have different actions taken on them at index and query time. The *_ssim fields are strings that are stored as-is, so Solr doesn't break the original string down into stems, doesn't make lower case versions, etc. You'll only get results on exact matches, thus they're good for facet fields and the like. *_tesim fiels, on the other hand, are text fields that have a lot of work done on them at index and query time, so that's why you get partial matches there. You'll just need to pick another version of the creator field from your schema if there is one (not sure how indexing works in your system) or maybe set up a copyfield.