While that's true, I have a vague memory that ≈350nm Pentium IIs also used broadly similar amounts of energy per instruction, maybe an order of magnitude more. But I might be misremembering that. I remember that you have much deeper hardware knowledge than I do; do you know of a better comparison table?
I don't think fair comparisons are really possible across different processes. There's a lot of interaction between silicon process, architecture, and workload.
The fairest comparison is between chips from major vendors in the same year. If you're just doing FP32 matrix multiplies, a 2026 nVidia GPU is ~1 pJ / FLOP, an Apple M4 is ~30 pJ, and a 2026 Intel Xeon is ~50 pJ. The quality of process and optimization is a little different between those 3 companies, but not by as much as 2x. But from those numbers, you can reliably conclude that the GPU architecture is much better for matrix multiplies.