Riley Yung

3.7K posts

Riley Yung banner
Riley Yung

Riley Yung

@karkwonk

I run a bug instagram. Former SpaceX, Commonwealth Fusion, 2nd grade coach’s pitch mvp

가입일 Ocak 2016
95 팔로잉109 팔로워
Hazel Appleyard
Hazel Appleyard@HazelAppleyard·
$10 per push-up or $1 million?
Hazel Appleyard tweet media
English
769
33
4.5K
1.9M
Riley Yung
Riley Yung@karkwonk·
@Robotbeat Granted I grew up in a small town, but by age 11 I was certainly allowed to bike anywhere in town on my own (3-4 mile radius). Hell it was a mile just walking to school in 3rd grade
English
0
0
1
31
Rob
Rob@rugbyyob·
@KJdaMAN11pogo @Gibhunter @30SecondYou @HazelAppleyard 50 pushs 4 times a day (can be done in minutes) = $2000 a day or $730,000 a year. 3 years in you would be on $2.2m. No way a $1m gets that high that quickly even with 20% annual return for 3 years.
English
1
0
1
152
flntwd
flntwd@flntwd·
@raines1220 Good point. It’s possible there are many more robotaxis in operation than currently reported.
English
1
0
3
405
Raines
Raines@raines1220·
Robotaxi Tracker has only 89 riders reporting their rides in Austin, and they’ve logged just 8,152 miles in total. Meanwhile, Austin Robotaxi has probably accumulated 1–2 million miles, meaning only about ~0.5% of the miles have been reported. From that ~0.5% of reported miles, 19 unique unsupervised Robotaxis have been discovered in Austin, and people are laughing about it. Well, to be honest, I’m laughing at you. Imagine I deploy 10 people to take Uber every day and count how many unique cars they discover. On the first day, they report discovering 10 cars. On the second day, they report another 10 cars (20 in total) Several days later, they report that 100 unique cars have been discovered in total (a pretty small number, right?). Then I can finally make fun of Uber: “Look! Uber only has 100 cars. How pathetic!”
Raines tweet media
English
12
5
116
9.4K
Riley Yung
Riley Yung@karkwonk·
If there were many more cars than riders reporting, you would expect a new car to be discovered almost every trip (like your uber example). But based on the mileage, the 89 riders have an avg of 10 rides each. Which means of the 19 cars discovered, they have each been reported an average of ~46 times. Very unlikely we are missing many (if any) cars.
English
0
0
2
238
Riley Yung
Riley Yung@karkwonk·
@quijaking @KyleJamesOlson @Gzalzi @YuletideFrost Ofc a few people do actually jump in front of a train, but people are dying all the time. If this experiment was run and the same number of people died as jump in front of trains on a daily basis, that would unfortunately just be a normal day
English
0
0
0
6
Riley Yung
Riley Yung@karkwonk·
Clearly from the polls/discussion there is not a reality where 8 billion people pick red. But as a red-instinct person, that is actually little surprising. There are plenty of things that ~all 8 billion people don't do, simply because we are not that dumb. ~All 8 billion people don't jump in front of a train. All 8 billion people don't walk naked into the sahara desert at noon. All 8 billion people don't eat raw roadkill. My first instinct upon reading the question is that not pressing blue is equally as obvious as the examples above, so actually no one is going to do it. Again, it is clear that is wrong, but you can perhaps understand why this is exasperating to red voters
English
1
1
1
112
Riley Yung
Riley Yung@karkwonk·
@catehall Almost certainly there are lower tranches with partial payouts
English
0
0
1
79
Freeman⚓️💪🏽
Freeman⚓️💪🏽@Emmachuuks·
What is that one exercise that hits all muscle groups?
English
61
6
116
17.5K
Riley Yung
Riley Yung@karkwonk·
@KenKirtland17 It's probably: 1. A tiered payout structure with that being the final tier/tranche (that's how the tesla ones were I believe) 2. Partially a marketing stunt for the IPO
English
1
0
13
513
Robotbeat🗽 ➐
Robotbeat🗽 ➐@Robotbeat·
@KenKirtland17 1) he says he wants to live to 100 (neither more nor less). 2) mortality is a great motivator. 3) it’s less ambitious than the 100TW of compute stipulation.
English
4
2
42
1.3K
Your mom’s favorite
Your mom’s favorite@favoritodetumai·
Really rubs me off the wrong way how quick people are to dismiss or downplay what Leila does just because she has no kids (look at the comments). An insane thing to to do to someone who’s a CEO of a $200M company
Dean Turner@DeanTTraining

Despite working 10-12 hours a day… Leila Hormozi: - Works out daily - Logged her food every day for 11 years straight - Is in better shape than 99% of women - Makes her man dinner every night Ladies, What’s stopping you from doing all this? Note: “Too busy” is NOT an excuse

English
8
0
7
2.5K
Riley Yung
Riley Yung@karkwonk·
@robinsonmeyer Are you suggesting that if you’re surprised to the downside, that data is not useful?
English
0
0
1
176
Robinson Meyer
Robinson Meyer@robinsonmeyer·
Once you get past basic mid-workout Strava HR/GPS/steps tracking, I’m increasingly unsure what the point of these devices are. Your sleep score can only surprise you on the downside. Your recovery score is not a reliable guide to injury risk. So what are we doing here?
Aakash Gupta@aakashgupta

You check your Apple Watch in the morning. Sleep score: 62. You decide it's going to be a foggy day. And then it is. A 2014 Colorado College study suggests the score itself causes the fog. 164 people walked into a lab. Researchers hooked them up to fake EEG equipment and told them the readout would show their REM percentage from the night before. Then they fabricated a number. Half the room was told 28.7%. Half was told 16.2%. The machine wasn't measuring anything. Participants took four cognitive tests. The Paced Auditory Serial Addition Test, where you add numbers spoken at increasing speed and hold your last sum in working memory while computing the next. And the Controlled Oral Word Association Task, where you generate as many words as you can starting with a single letter under time pressure. Both are gold-standard measures of attention and executive function used in clinical neurology. The 28.7% group outperformed the 16.2% group on both. Significantly. How rested participants actually felt that morning predicted nothing. The mechanism is mindset priming an executive resource. When you believe you slept well, you allocate cognitive effort more aggressively. You don't conserve. You don't pre-disengage. Belief about the resource changes how you spend it. Two control conditions ruled out demand characteristics. Participants weren't trying harder because they thought they should. Real measurable cognitive performance shifted with the number on the readout. The Apple Watch sleep score. The Oura ring readiness number. The morning ritual of checking either one is taxing the resource you're about to need. The performance gap from a fabricated REM percentage was larger than the gap from how rested participants actually felt. The number was louder than the night.

English
27
12
423
75.1K
Riley Yung 리트윗함
Brooks Otterlake
Brooks Otterlake@i_zzzzzz·
This is just like being alive in the 1600s when they got good at making complicated clocks and deduced that every complicated thing in the universe probably functioned exactly like a clock
Dwarkesh Patel@dwarkesh_sp

There's a quadrillion-dollar question at the heart of AI: Why are humans so much more sample efficient compared to LLM? There are three possible answers: 1. Architecture and hyperparameters (aka transformer vs whatever ‘algo’ cortical columns are implementing) 2. Learning rule (backprop vs whatever brain is doing) 3. Reward function @AdamMarblestone believes the answer is the reward function. ML likes to use pretty simple loss functions, like cross-entropy. These are easy to work with. But they might be too simple for sample-efficient learning. Adam thinks that, in humans, the large number of highly specialised cells in the ‘lizard brain’ might actually be encoding information for sophisticated loss functions, used for ‘training’ in the more sophisticated areas like the cortex and amygdala. Like: the human genome is barely 3 gigabytes (compare that to the TBs of parameters that encode frontier LLM weights). So how can it include all the information necessary to build highly intelligent learners? Well, if the key to sample-efficient learning resides in the loss function, even very complicated loss functions can still be expressed in a couple hundred lines of Python code.

English
105
1K
13K
787K
Riley Yung
Riley Yung@karkwonk·
@Credib1eGuy Many things have been found with essentially no tradeoffs relative to the benefits. Even ones tinkering with the human body. Polio vaccine (and later many virus vaccines), antibiotics, brushing your teeth…
English
0
0
0
23
Guy
Guy@Credib1eGuy·
All this peptide and retratide whatever stuff, GLP, ozempic etc I feel like it’s gonna come out with some crazy bad side effects or consequences in 10 years It’s just common sense nothing in life has no trade offs
English
300
23
1.1K
514.4K
Riley Yung
Riley Yung@karkwonk·
@Logo_Daedalus This is true but may I note that your best theory of mind for red voters is that they live “a life of aimless stateless blah blah” and would be borderline excited to die in order to actually feel something
English
1
0
1
65
R.Сам 🦋🐏
R.Сам 🦋🐏@Logo_Daedalus·
I think a lot of morality failures are theory of mind failures— thus making revealed preference being truly the revealed revealed preferences of others. “Surely my best friend is a high iq red voter— wait, what the fuck? He voted blue?” Etc
English
4
3
214
9.2K
R.Сам 🦋🐏
R.Сам 🦋🐏@Logo_Daedalus·
All these people suggest “in real life more people would vote red” would be so fucking surprised to discover that blue would have an even larger majority if death was *really* on the line.
English
139
81
3.4K
61.4K