_akhe

31 ms·

301.
by _akhe 2y ago
I meant $100M in annual revenues :) Your point is taken though, I think a lot of VCs would scoff at our present-day numbers despite our growth and outlook. It's very much still a small business, yet by far the most impressive thing eit
302.
by _akhe 2y ago
> the smartest people happen to go to Stanford There's no way this is even possible, let alone true. This might be a hot take but if you're under the age of 25, it's likely that you know very few things total. And the thin
303.
by _akhe 2y ago
I really appreciate this comment. Service offered / payment received is irreplaceably wholesome and satisfying :) especially when they come back as repeat customers or become regular clients. Repeat business and gradual growth gives us
304.
by _akhe 2y ago
Bookish, yeah, the look has aged well. My intuition was that there are more overly-confident types who think they want to be a founder rather than those not confident enough to do it, but after reading some comments here maybe my intuitio
305.
by _akhe 2y ago
One of the reasons I like this site is that it feels slightly academic in a way that a competing site like Reddit does not. And being a somewhat exclusive post-academic institution is not a bad image or message IMO. Maybe there are more pro
306.
by _akhe 2y ago
Re: "won't get in" How do I balance this encouraging advice with everything I've heard online about YC basically being a post-Ivy League thing? Seems like there's basically a 0% chance a random person would get into
307.
by _akhe 2y ago
Dennett on Closer To Truth (just uploaded) https://www.youtube.com/watch?v=IHvIXe4DOEc
308.
by _akhe 2y ago
Dennett is a mainstream eliminitavist, his appeal is that he removes extra fluff around philosophical concepts like consciousness and the brain. In a world where half the consciousness gurus are talking about unproven quantum stuff, souls,
309.
by _akhe 2y ago
> I wonder if we could continue to get insights from him I think so. His views were somewhat rigid and materialist. As a de-fluffer, he's great, and I'm sure he will continue to be quoted especially as we make progress on AI an
310.
by _akhe 2y ago
First experience with llama3 (via ollama) 1) It keeps unnecessarily hash-tagging sentences, but listens better than mistral at removing them if I specify 2) It failed my pop culture "Arthas test": Here's llama3: >>&
311.
by _akhe 2y ago
Really cool read, I liked seeing all the examples. I wonder if it would be beneficial to train on lots of static views of the character too - not just the frames - so that permanent features like the face gets learned as a chunk of adjacent
312.
by _akhe 2y ago
> China suppresses citizens while EU protects citizens! Lol this is the real silly provocative comparison. China bans sites & apps from the West that violate their laws - the ad tracking, monitoring, censorship & influencer&#x2
313.
by _akhe 2y ago
This would have been cool for hardware wallets when Ethereum was relevant.
314.
by _akhe 2y ago
Their UI components are free lol I use Tailwind UI (the React components) all the time in projects, never paid for anything. It's a great place to start for a design system and component library in a new project. After checking again I
315.
by _akhe 2y ago
What's the second axiom?
316.
by _akhe 2y ago
ollama pull llama3 It's working :)
317.
by _akhe 2y ago
This is comment is cooler than my Arctic Vault badge on GitHub. Who were the countless unknown contemporaries of Giotto and Cimabue? Of Da Vinci and Michelangelo? Most of what we know about Renaissance art comes from 1 guy - Giorgio Vasari.
318.
by _akhe 2y ago
> I use a custom stylesheet and tune them to be more readable See this is what I'm missing! To me it invokes laughter for some reason - like the message is sailing off into the fog, bye! I'm like "what's that message
319.
by _akhe 2y ago
I always assumed shills were like paid posts (HN's monetization). Unless you mean like non-startup related shills? Agree, overall HN is pretty solid. There's basically no bullshit, yet at first glance you can't really tell it
320.
by _akhe 2y ago
I appreciate the honesty of this post - I feel almost exactly the same way about most things you said. Social media seems over-engineered, and it's interesting that even behind the simple look of this UI there are layers of complexity,
321.
by _akhe 2y ago
> mature audience > seem to genuinely want to have a discussion around its merits So true, and I think that's what overall makes HN worth it in a way that Discord or Reddit are ultimately not. Heavy moderator culture tends to ren
322.
by _akhe 2y ago
Great observations regarding personal/back-and-forth, I skip those too as a reader yet get caught up in it sometimes when I post a top-level thing. > weak signal of popular consensus Basically what I've been using it for recent
323.
by _akhe 2y ago
Your understanding of it seems far less stressful than mine! I like the idea of voting on conversational quality rather than agree/disagree. Sometimes disagreement really just means surprised or offended, and sometimes those are conver
324.
by _akhe 2y ago
> as someone with a PhD in the topic, who has taught many students, and published dozens of papers :joy: How about a repo I can run that proves you know what you're talking about? :D Just put together examples and you won't hav
325.
by _akhe 2y ago
> We now have a method of embedding a variable length piece of text into a fixed size vector Question: Is it a rule that the embedding vector must be higher dimensional than the source text? Ideally 1 token -> a 1000+ length vector? T
326.
by _akhe 2y ago
Very nice, thanks for sharing! I noticed their token sequence length is fixed: Constructing 7-gram LM ... Created trie with 21,502,513 nodes. People on here will be happy to say that I do a similar thing, however my sequence length
327.
by _akhe 2y ago
Lol so is Tailwind
328.
by _akhe 2y ago
This is a simple example where it scores their frequency. If you scored every word by their frequency only you might have embeddings like this: act: [0.1] as: [0.4] at: [0.3] ... That's a very simple 1D embedding, and li
329.
by _akhe 2y ago
I was referring to the observational proof when I said that - the entire Higgs craze was about the proof - not about math as both of you guys are saying. The math part was thought of in the 60s, 2012 was about the actual discovery - did you
330.
by _akhe 2y ago
> "math for Pluto's rough position came first and directed observers (roughly) where to look" But the math wasn't proof that Pluto existed, it wasn't proven until it was discovered.

More ›