brokensegue

We could hit the entire planet with nukes. Yeah maybe <1% would survive immediately, but who knows about long term. I think that's close enough.

I think better than priors would be a closed loop where you tell it what the right answer was (or some signal) and they monitor and fine-tune for you

give it an ascii representation \j

They also have custom brain bits evolved to aid in this task

you can ask structured output for probabilities...not that they necessarily mean anything.

Multiple places

Aren't the new models thinking in neuralese?

surely you'd have to also control for the kind of driving rideshares do (e.g. local v. highway) to have any real sense.

maybe we could look for insurance rates?

Isn't storj a crypto thing? I don't think they'd want to be involved. I get the sense they are idealists

Also it includes insurance and such that doesn't appear in TC

Disclaimer: I work for the wmf

If you're doing a job you aren't unemployed. Underemployment is measured separately

IQ tests? Very good. But most are in the dataset so it's not very meaningful

Do you want your phone to allocate 4GB to read a mailing list?

yeah that was my understanding. for some reason corn syrup is vilified. the real problem there (as I understand it) is corn subsidies.

Is there evidence supporting the idea that sugar is more satiating? My understanding was that they were equally unhealthy

Asianometry is underrated. So much in depth content

it's cool it lasted so long

Per hour?

But surely people's willingness to listen to an AI is contingent on their past experience with this model

Why didn't they use a model people actually use?

Fable told me

> Verdict: I checked every step and found no error. The argument appears to be a correct proof of the Cycle Double Cover conjecture, modulo two standard cited results (the reduction to loopless cubic graphs and the Jaeger–Kilpatrick 8-flow theorem, both real and well-established).

> Two caveats: this would settle a ~50-year-old open problem in three pages, so it deserves independent expert scrutiny regardless of my check; and I couldn't reach the web from here to confirm the paper's provenance or any community response, so I can't tell you its status beyond the mathematics itself.

personally, i've never had good luck with MT's quality.

Is this at all related to kicanvas.org

Brainfuck is probably the most famous esoteric programming language. It's mentioned (and used) in published papers e.g. https://link.springer.com/article/10.1007/s10994-019-05821-2. Probably not a good comparison point.

I think a more comparable language would be https://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletio...

if people want there to be an article on Odin they can contribute to the current draft. https://en.wikipedia.org/wiki/Draft:Odin_(programming_langua...

Websites can request to be removed. The data is still kept though

it's really not much to manage assuming you are already doing backups. you ~ pay the energy cost either way.

You mean neither?

there are models between 2-grams and 600m param models that would be good options. i don't expect a 2-gram to do very well here. also i'm not sure why this model isn't a fine choice if it solves their problem

you should link your data to wikidata which will get you free connection back to crunchbase and other sources e.g. https://www.wikidata.org/wiki/Q97041185

You could even back some of the data from there

EU tried to ban tracking cookies. Anyone could've seen the result coming (every site now disrupts your reading). The defeatist attitude would've been right there.

I don't see how games won't just charge you a $0.01 "subscription" that lasts 5 years or various other such sidesteps. it'll just make everything more annoying.