The term GPU wasn’t used yet. It got applied as something of a marketing term to cards that had hardware transform and lighting, and that was indeed the GeForce 256. Before then, they were “3d accelerators”.

You can see this on the Wiki page for the GeForce: en.wikipedia.org/wiki/GeForce_256#Architecture

GeForce 256 was marketed as “the world’s first ‘GPU’, or Graphics Processing Unit”, a term Nvidia defined at the time as “a single-chip processor with integrated transform, lighting, triangle setup/clipping, and rendering engines that is capable of processing a minimum of 10 million polygons per second”.

So it kinda depends on perspective. If you take Nvidia’s marketing at face value, then the GeForce 256 was, indeed, the first GPU. You could retroactively apply it to earlier 3d accelerators, including the SNES Super FX chip, but none of them used the term at the time.