tikkun

45 ms·

601.
by tikkun 3y ago
When do you (HN readers) think that we'll have an open source model that scores 1150 or higher, and where do you think it'll come from?
602.
by tikkun 3y ago
A list, ranked by person/sponsor, ranked by total value reduction across all SPACs they did since some time point (merger? before then?), and normalized for multiple compression by looking at the ratio of some overall public tech index
603.
by tikkun 3y ago
Yes. I feel there should be some kind of anti-list of people who dumped SPACs and ICOs on the public. I personally feel it was pretty gross behavior (in many, but not all, cases), and the people who did the egregious ones mostly knew at the
604.
by tikkun 3y ago
It sounds like you kinda don't want to go to university but you feel like you should. Depending on how old you are, you could: * Assume you won't go, and work on whatever you want to work on in the meantime * Apply anyway, and con
605.
by tikkun 3y ago
For me: I only do the "activities that I can't help but do" or the "goals that I can't not reach" The inputs that I can't not do, and the outputs that I can't not have. This is a good way to simplify
606.
by tikkun 3y ago
1. Claude by Anthropic (but, expect to wait weeks/months before getting off the waitlist, which makes their offering pretty uncompelling) 2. Cohere 3. https://inferkit.com or https://textsynth.com/pricing.ht
607.
by tikkun 3y ago
Would be neat if it had stable diffusion powered image generation, to help people visualize and remember their passwords.
608.
by tikkun 3y ago
It would've been smart for them to do this transition during the COVID stimulus era - lockdowns, work from home, stimulus checks, low interest rates.
609.
by tikkun 3y ago
LLM for technical audiences, and AI for non-technical audiences, I think.
610.
by tikkun 3y ago
Everything can be viewed as a product. A scientific paper is a product. Products need distribution and sales. Talks and 1-1 conversations are "sales" for papers.
611.
by tikkun 3y ago
tim_sw: Thanks for all the interesting things you submit to HN, I recognize your username from a few articles I've upvoted.
612.
by tikkun 3y ago
> I've been thinking about interviewing for jobs that I don't plan on getting, just to get a better offer and force my boss hand, does that sound like a decent idea? Yes. Get the better offer. But you don't have to force t
613.
by tikkun 3y ago
I love finding useful strategies and tactics from nature, and applying them in my life and work. Another example: Redundancy. We have two lungs, two nostrils, etc. Anything that I'm stressed about breaking or not having enough of, if i
614.
by tikkun 3y ago
Related: "Here's a phenomenon I was surprised to find: you'll go to talks, and hear various words, whose definitions you're not so sure about. At some point you'll be able to make a sentence using those words; you w
615.
by tikkun 3y ago
I'd prefer non-SaaS companies find a clearer way to represent it. I agree that that's how SaaS companies use it, and I think it makes sense in SaaS (where the gross margins are almost always 80%+). When someone uses a number aimed
616.
by tikkun 3y ago
"from ground zero to $1M+ Annual Recurring Revenue (ARR) and 25 employees." If it's $1m+ in recurring net revenue (i.e. after the cost of the staff provided to customers) then I think it's right to say $1m+, otherwise, i
617.
by tikkun 3y ago
I think Bryan Johnson's approaches don't appeal to me, but I'm really glad that he's doing weird things! I think a world where more people can do uncommon experiments is a good one.
618.
by tikkun 3y ago
It's not a randomized control comparing 1 vs 2, it's the claim that one may be better because in developing countries it's more practical to have people do one.
619.
by tikkun 3y ago
> Is it because on HN, people are after dev tools, not studying tools? Pretty much. It's mostly the wrong audience. HN is interested in things that are either aimed at a technical audience, or where you can show some really unique w
620.
by tikkun 3y ago
There are some "drag and drop" type solutions, like https://www.chatbase.co/ . There are various more - search for custom chatgpt on product hunt and you'll find a lot.
621.
by tikkun 3y ago
My understanding is that the answer is generally: not yet. (I wish, I suspect we'll be able to learn some interesting things about the universe, about humans, and so on, by seeing what LLMs found to be highly explanatory / high or
622.
by tikkun 3y ago
> Are there papers that peruse what kind of concepts the model is actually building/learning in those heads and layers? > There are large teams who spend months tuning those models. Do those teams have access to those internal co
623.
by tikkun 3y ago
Someone mentioned a rumor about "Apple has achieved 1:1 3d video chat where there avatars are almost completely photo realistic" https://news.ycombinator.com/item?id=35995072 Anyone know a source link?
624.
by tikkun 3y ago
It would be great to see a set of examples - before, after.
625.
by tikkun 3y ago
Related: "Almost all founders learn brutal lessons during the first year, but some learn them much more quickly. Obviously those founders are more likely to succeed. So it could be a useful heuristic to ask, say 6 to 12 months in, &quo
626.
by tikkun 3y ago
For anyone curious, the ESP box mentioned is likely: https://news.ycombinator.com/item?id=35948462
627.
by tikkun 3y ago
https://www.reddit.com/r/LocalLLaMA/wiki/models/ https://chat.lmsys.org/?arena (Then click Leaderboard)
628.
by tikkun 3y ago
Looks nice. What are you using for the GPU hosting – a serverless provider like Modal/Banana/Replicate, or a non-serverless provider like Vast or LambdaLabs or AWS GPU instances?
629.
by tikkun 3y ago
Has anyone built the "online photoshop that incorporates all of the latest AI image editing tools asap and sells access as a premium subscription with lots of GPU access for smooth editing" business yet? I'd be curious to kno
630.
by tikkun 3y ago
I've run into similar issues. My solution is I added signs, and delivery instructions that reference the signs.

More ›