The TOPS numbers from the wiki page seem wrong. TPUv1 had 92 TOPS (uint8); for TPUv3 the “90 TOPS” refers to a single chip, but I’m fairly sure that when the paper says “8 TPUv3s” they mean 8 cards, as that’s how they are available on Google Cloud (1 card = 4 chips).
The TOPS numbers from the wiki page seem wrong. TPUv1 had 92 TOPS (uint8); for TPUv3 the “90 TOPS” refers to a single chip, but I’m fairly sure that when the paper says “8 TPUv3s” they mean 8 cards, as that’s how they are available on Google Cloud (1 card = 4 chips).
Huh, thanks! I guess my guesstimate is wrong then. So should I multiply everything by 8?