And that field is a long text.

Go here: https://lzstring.azurewebsites.net/?handler=ConvertLzStringCompress put in your input string and look at the output. When I entered "this was a really verbose sentence and I should really compress it" it output (in the first box): "஀낖ೠЎ펔Ӡ匸ۃӲ胝噀⌁䩲ζᔪذ긪IJ聉䩁⁗೛䦇ᑺ␂�獁ᨘ恀" Note the "�" towards the end. I suspect it's the way it's rendered, the character is not part of the character encoding set. Maybe you need to choose a different setting or further encode the output, check the other boxes on that page for one that's sane.

It looks like it works fine with

compressToUTF16

It wasn't just rendering. If you ignored how it looks, and made a script to save it, fetch it, then decompress it, it still will find the input to be corrupted. btw, how did you find that link? I can't find it here https://pieroxy.net/blog/pages/lz-string/index.html

Google 🤣

Thanks ☕