yifanl

12 ms·

211.
by yifanl 2y ago
To me it seems like it comes from experience solving sudokus. If you've ever tried it (or watch Cracking the Cryptic if you haven't), you don't store just the solution of each square, you keep track of what you've ensure
212.
by yifanl 2y ago
Being confident enough to let people think you aren't fashionable is a statement on its own :P
213.
by yifanl 2y ago
I think someone being opinionated about their fashion sense is a much more powerful message than at most choosing which colour of plastic (and often we can't even do that!). edit: and certainly, there are people who think the apple wat
214.
by yifanl 2y ago
Have books of any form ever been categorized as sexy?
215.
by yifanl 2y ago
Is there still evidence that more compute = better model?
216.
by yifanl 2y ago
The usage of existing but cheaper nvidia chips to make models of similar quality is the main takeaway. It'll be much harder to convince people to buy the latest and greatest with this out there.
217.
by yifanl 2y ago
It does, but proving that it can be done with cheaper (and more importantly for NVidia), lower margin chips breaks the spell that NVidia will just be eating everybody's lunch until the end of time.
218.
by yifanl 2y ago
It does at the moment. At some point in the future, it might not rank, at which point he and anyone else devoted to the rat race will get their teams to replace their thumbnails with whatever new thing gets more clicks.
219.
by yifanl 2y ago
I would suspect if someone tried to ship a full windows OS on the Google play store, that would also be banned for trying to break the app sandbox.
220.
by yifanl 2y ago
People don't understand that LLMs are just statistical magic, by design of the LLM providers because LLMs wouldn't take off if people understood they are just statistical magic.
221.
by yifanl 2y ago
Is it just an open secret that the models are currently just being hardcoded for random benchmarks? Seems weird that people would be asking Putnam problems to a chatbot :/
222.
by yifanl 2y ago
Sure, but does it make sense that your cubicle row of data entry guys are on the cutting edge knowledge of LLM expertise? Because the way LLMs are being pushed in every spreedsheet software would imply that's something someone is askin
223.
by yifanl 2y ago
She releases a video like, once a year, if that. We can have a little trolling, as a treat. And I would suggest she's able to go on and on about these particular experiences for so long is that they're genuinely unique experiences
224.
by yifanl 2y ago
And in the more recent past, we asked Juicero if squeezing a small volume of bags by hand could be done more cheaply by machines, and they told us "absolutely!"
225.
by yifanl 2y ago
I don't recall anywhere in her video where she didn't bring up expectations without also bringing up an example of when the Disney company had already implemented the idea.
226.
by yifanl 2y ago
For accuracy, it's a 100 page guide to technical writing, followed by a 50 page sample rulebook for Stardew Valley implementing all the ideas :)
227.
by yifanl 2y ago
Board game rulebooks are documents that need to cater to multiple audiences including both new players learning from the rulebook and experienced players looking to skim through for very specific interactions they have forgotten. They also
228.
by yifanl 2y ago
Google's moat with this current wave of AI is pretty obvious: They own the compute resources inhouse. Apple isn't immediately seeking to compete in this field, presumably because they don't see a deep enough moat.
229.
by yifanl 2y ago
I'm largely uninterested in the specific name you want to give it and more if its worthy of punishment.
230.
by yifanl 2y ago
There's nuances to the amount of harm dealt to the authors based on what sources you are stealing from, but it's irrelevant here, as the specific incident we're talking about is whether or not the student is the actual author
231.
by yifanl 2y ago
If you write in a manner that gets you dismissed as a chatbot, then you've still failed to communicate, even if you physically typed the characters in the keyboard. The essence of communication isn't how nice the handwriting is, i
232.
by yifanl 2y ago
Which would slow down user uptake, I imagine.
233.
by yifanl 2y ago
That would only be meaningful if ChatGPT made as much money as Google did per query (as is, I believe they lose money on each, so they can hardly make it up in volume).
234.
by yifanl 2y ago
Because we had an implicit understanding that it would continue to be humans reading the code.
235.
by yifanl 2y ago
The same thing Amazon would have to do to detract fake listings: Add friction. Of course, even mentioning the f-word is forbidden, so...
236.
by yifanl 2y ago
I know what the words means to me, I don't know what it means to Matt, and his opinion seems to be the relevant one.
237.
by yifanl 2y ago
It's not obvious that in the general case, employees can be motivated by getting work done if the reward is more work to do.
238.
by yifanl 2y ago
It's not obvious to me that employees are motivated to be more productive for the sake of productivity.
239.
by yifanl 2y ago
I mean, what's the claim then, there's no such thing as an illegal idea? You can't assign a semantic value to a legal system.
240.
by yifanl 2y ago
Is the claim there some special property that makes it impossible to convey hate as opposed any other type of idea through text? That seems extremely wrong, especially in this context, given that LLMs make no attempt to formalize "idea

More ›