aabhay

14 ms·

181.
by aabhay 1y ago
For any interested, the best book I found on this topic is Project Japan ( https://a.co/d/fcrNp6p ). It dives into the history of the whole Metabolism movement (in my opinion, an effort to create more modular and dynamic
182.
by aabhay 1y ago
Lmao, gentrify cracked me up
183.
by aabhay 1y ago
Does anyone use OpenSearch for its knn and vector capabilities? Is it any good? It’s always hard to know with systems like this whether it works at scale until your team is fighting fires.
184.
by aabhay 1y ago
I think the better argument is about the direction of change versus the current magnitude. If we are to believe that the models will get bigger, use more tokens, work for longer, this calculation can easily become very very skewed in the ot
185.
by aabhay 1y ago
But you often want a number that’s more correlated with the average. Imagine a distribution that has 80% very low values and 20% insanely high values. The median won’t capture changes in that 20% region at all. So you could see numbers not
186.
by aabhay 1y ago
I personally think the mere idea that you can map on abstract functions that have infinities and boundary conditions, on to a representation that extremely precisely models our physical world, is more astonishing than having a low level und
187.
by aabhay 2y ago
I’m surprised that the world’s greatest universities are in the United States. Why weren’t other countries providing funds?
188.
by aabhay 2y ago
On the other hand, if you think of a programming language as a specialized tool then you choose the tool for the job and don’t reach for your swiss army knife to chop down a tree. The problem with gradually typed languages is that there are
189.
by aabhay 2y ago
Curious if you have any examples since I didn’t read any zeal from the posted essay.
190.
by aabhay 2y ago
I think this might be true in an invoicing based world but I agree that PoS is a misnomer. But I can imagine e.g. a field sales rep in construction taking orders on site in a rural area then syncing up when they return to connectivity.
191.
by aabhay 2y ago
OpenAI was started with the express goal of undermining Google's potential lead in AI. The fact that they time launches to Google launches to me indicates they still see this as a meaningful risk. And with this launch in particular I f
192.
by aabhay 2y ago
Generally, no, as a principle. You don’t want to learn things in chronological order, you want to learn them in intrinsic ness order, to the extent it is feasible. If you really want to learn computing, start from scratch. Go from logic gat
193.
by aabhay 2y ago
Triton is not that bad at all, considering the wide scope of systems it has to support (tensorrt, onnx, multiple generations of pytorch, cuda, python). It was much nicer than the old Torchserve project which was JVM based.
194.
by aabhay 2y ago
This is very narrowly focused on LLMs, whereas triton is still useful for running all kinds of ML models. In practice, Triton is a very poor choice for LLMs specifically because it has none of the required non negotiable features like KV ca
195.
by aabhay 2y ago
We use discord as a workplace chat platform that has allowed us to make great use of its converged “internal + external” structure. Basically, its works as both a team commons as well as a customer support and customer champion platform. Wi
196.
by aabhay 2y ago
While I agree with your view point, I think the celebrity mathematician status of Terrence Tao made it obvious to me that the title was referring to mathematical solutions e.g. solving a system of equations.
197.
by aabhay 2y ago
It’s always the popular clubs that make the most rules
198.
by aabhay 2y ago
Making good on the username, for sure!
199.
by aabhay 2y ago
> Makes a bunch of money > tries to be Elon Musk > Breaks up with girlfriend > joins DOGE (to not be like Elon…?) > Mentions repeatedly the desire to not be cringe Sounds like a cringey Elon fantasy despite the self awareness
200.
by aabhay 2y ago
What systems would you recommend for larger deployments?
201.
by aabhay 2y ago
> Both companies had arrived at a similar view of the future where AI will redefine every business application and workflow — and reinvent productivity as we know it today into a place where humans and AI work together everywhere you get
202.
by aabhay 2y ago
Would create too much frenemy energy with microsoft mothership
203.
by aabhay 2y ago
This work doesn’t convince me that alignment faking is a concern. Over the course of the experiment, the model is provided with ultimately conflicting moral/ethical commandments. In the case of such conflicts there are no purely optima
204.
by aabhay 2y ago
Disagree. Stanford researchers have been making datasets for years and years. SQUAD is another Stanford dataset. Everyone knows that publishing datasets gets you citations. But that gig is now sorta over because the word is out.
205.
by aabhay 2y ago
Hard to believe that ImageNet fueled advancements in that field. Dataset creation is a very widely known “citation hack” because it forces others to cite your work even though the dataset is often used for thrift reasons over any genuine va
206.
by aabhay 2y ago
She has been exceptionally good at creating a narrative IMO. I agree that her actual accomplishments are less clear. At Stanford, a person in her position gets added as last author to all her students’ work so her citation count is very ske
207.
by aabhay 2y ago
My description of the power set is by definition allowing all to imply some
208.
by aabhay 2y ago
The puzzle is referring to the concept of a vacuous truth ( https://en.wikipedia.org/wiki/Vacuous_truth ). In most logic frameworks, the All function (upside down A in standard logic notation) is true if and only if no s
209.
by aabhay 2y ago
Biology has an incredible set of priors for this, given that actual neurons have been experimenting with topologies for millions of years. The neuron itself can be thought of as a meta-topology in that it constructs potentially arbitrary hy
210.
by aabhay 2y ago
“Hundreds of artists provide unpaid labor through bug testing, feedback and experimental work for the program for a $150B valued company” This is a hilarious basis for protest

More ›