alt Hacker NewsThe repo has fp16 numbers.
Also this one for 4070s shows that its for fp8 as well:
https://github.com/JustAResearcher/gpu-unlock-ada/releases/t...
The repo has fp16 numbers.
Also this one for 4070s shows that its for fp8 as well:
https://github.com/JustAResearcher/gpu-unlock-ada/releases/t...
it's the widening (for example: fp16 * fp16 = fp32) instructions, which are useful on some loads.