GPU: NVIDIA GeForce RTX 4060 Ti, sm_89, PCI <target>
FP16/FP32: m=256 n=256 k=256 seed=1, 65536 cells, failures=0, max_abs=0
BF16/FP32: m=256 n=256 k=256 seed=1, 65536 cells, failures=0, max_abs=0
FP16/FP32: m=64 n=96 k=128 seed=2, 6144 cells, failures=0, max_abs=0
BF16/FP32: m=64 n=96 k=128 seed=2, 6144 cells, failures=0, max_abs=0
FP16/FP32: m=127 n=65 k=95 seed=3, 8255 cells, failures=0, max_abs=0
BF16/FP32: m=127 n=65 k=95 seed=3, 8255 cells, failures=0, max_abs=0
