They say ECC works (?) so you would know.

I recently had a faulty A100, for some reason the provider turned off ECC on that card, so it silently failed and gave junk results... Took me a minute to figure out.

▲

fn-mote a day ago | parent [-]

How did you ever figure this out?

	▲	sva_ a day ago \| parent [-]
		The weights loaded fine, but as I was training the loss made no sense. So I wrote some basic matmul test.