When the Z340 was decoded, some years ago now, I don’t think that anyone was surprised to learn that the ‘original artifact’ which was set into code would have been a piece of paper with some inane sentences written on it by hand and conveying nothing of interest at all. Of course, the actual artifact as a physical object was not recovered, but the text was.
We don’t have even this much currently for the Z32. That said, however, it is still possible to propose the ‘form’ of the original artifact in this case also. Essentially, but by way of example only, uncomplicated logic must have us believe it would have been nothing any-the-less concrete and tangible than this:

“This isn’t right! It’s not even wrong!”—Wolfgang Pauli (1900–1958)
[cont…]
Of course, when it came to the encoding in this latter case, you can’t ‘encode’ a physical map. So, what you do is you take the coordinates (bearing, range) of the marked location of your device and you encode that. Again, the process of obtaining the coordinates is nothing more complicated than an act of measurement, so:

“This isn’t right! It’s not even wrong!”—Wolfgang Pauli (1900–1958)
[cont…]
Having then composed the coordinates into a cipher you can’t, of course, just send this alone, so you have to send the original map as well – being sure, of course, that you first erase the site location ‘x’. You post this:

“This isn’t right! It’s not even wrong!”—Wolfgang Pauli (1900–1958)
[cont…]
… and, having done so, you can then sit back for the next 50 years and watch while legions of sleuths set out to demonstrate how you must have applied some method exponentially more intricate than simply this.
“This isn’t right! It’s not even wrong!”—Wolfgang Pauli (1900–1958)