erm... gradualist sisters... our uhm, response?
>>17043802Grok can compute remainders. Wow.Calculators on suicide watch.
>>17043802>it still randomly shits the bed>it would probably shit itself hard if you doubled the domain limitspowerful..
>>17043802Why would you train on 50% of the data? Unless it’s because the other 50% is redundant, in which case the title of the chart itself is spamlike and vacuous. Essentially, the title of the chart itself proves that you’re either a liar or an idiot lol.
>>17043808he doesn't mean grok>https://en.wikipedia.org/wiki/Grokking_(machine_learning)