ntonozzi

10 ms·

91.
by ntonozzi 3y ago
Why would that be annoying? It’s much easier to understand, predict and truncate appropriately than having to explain all of these different tokenization schemes to devs.
92.
by ntonozzi 3y ago
That article is so weird: > Of course, there is a selection bias in who comes to me. There are people out there doing just fine on 5000 units of Vitamin D daily. I only see the ones who develop high calcium levels. But I see enough of th
93.
by ntonozzi 3y ago
And toothbrushes too, did you know in some countries it is legal to brush your teeth and buy a toothbrush even if you didn’t see a dentist in the last year?
94.
by ntonozzi 3y ago
Not soon enough, the iOS transcription is terrible compared to even the small whisper model. I really hope Apple revamps voice recognition this fall, but I'm not putting money on it being Whisper quality.
95.
by ntonozzi 3y ago
Thanks for the helpful explanation. What is the context length in this explanation?
96.
by ntonozzi 3y ago
https://nutritionfacts.org/video/flashback-friday-benefits-o...
97.
by ntonozzi 4y ago
Nuclear waste from power plants is not a legitimate issue: https://thebreakthrough.org/issues/energy/the-boring-truth-a... , https://zionlights.substack.com/p/everything-i-believed-abou... .
98.
by ntonozzi 4y ago
He’s not saying that everyone had lower productivity through the pandemic, but that those hired during the pandemic and working remotely had much lower productivity.
99.
by ntonozzi 4y ago
Don't forget cars.
100.
by ntonozzi 4y ago
Check out https://github.com/erikbern/ann-benchmarks for some benchmarks on some of the different ANN libraries out there. I'd be interested in hearing other's experiences using these libraries in production.
101.
by ntonozzi 4y ago
Typically and hopefully they would have to return the money that they did not spend, and it's better for the investors to get that portion of their money back than to have founders run it into the ground.
102.
by ntonozzi 4y ago
Perhaps more carbon efficient, but 15x less energy efficient: > Biking takes around 25 kcal/km [iii] above basal metabolism, which is equivalent to .11 MJ/km. A typical car in the US gets 25 mpg, or 9.5L/100 km, which is
103.
by ntonozzi 4y ago
Llama is trained with _more_ data than is chinchilla optimal in order to make it better and cheaper at inference time, instead of just getting the highest quality of model that you can based on a given training budget. Llama has fewer param
104.
by ntonozzi 4y ago
If you only deposited $1000 they are not going to give you $250k back.
105.
by ntonozzi 4y ago
Designers sure love gradients+noise. And dang does it look good.
106.
Pinecone Partial Database Outage Postmortem(pinecone.io)
1 points by ntonozzi 4y ago|1 comments
107.
by ntonozzi 4y ago
My suspicion is that this is true for you, but untrue generally. When walking and biking, all of my close calls have been with large trucks. Drivers typically feel more confident going faster in large vehicles because they are higher up, an
108.
by ntonozzi 4y ago
How are they different from models that already exist like FLAN-T5?
109.
by ntonozzi 4y ago
Including the word 'phone' six times in a popular blog post is not going to help their predicament.
110.
by ntonozzi 4y ago
I think Flan-T5 is fast enough, but I don't think it generates text or abstract reasoning at nearly the same level as current GPT-3 models. This indicates a deficiency in the benchmarks and metrics that we use to evaluate LLMs. For gen
111.
by ntonozzi 4y ago
For the most part I agree, but Storj seems like it is at least attempting to solve a problem not created by blockchain: https://www.storj.io .
112.
by ntonozzi 4y ago
Amazon emplyees actually have a negative cash salary that they pay back with equity they sell.
113.
by ntonozzi 4y ago
Why do you say logistically no? The challenges of distributing the runtime?
114.
by ntonozzi 4y ago
Java, Go and C# (and node) have very similar performance, e.g. https://benchmarksgame-team.pages.debian.net/benchmarksgame/... . For all of them, the key to writing high performance code is avoiding allocations and boxi
115.
by ntonozzi 4y ago
Go belongs in the exact same bucket as Java and C#.
116.
by ntonozzi 4y ago
I agree MacOS feels dated, some examples: The Windows window manager is far better than the MacOS equivalent, the UI is higher quality, more consistent and more discoverable, and PowerShell is better than zsh.
117.
by ntonozzi 4y ago
Oops, good point! EFF also has some good articles about software patents: https://www.eff.org/issues/patents .
118.
by ntonozzi 4y ago
Can you explain why it's not what EFF means? To me, the part of software patents that doesn't make sense is that you can patent any possible combination of ideas — in this case the combination of using voice and remotely controlli
119.
by ntonozzi 4y ago
Software patents are a huge drag on small companies and innovative large ones, it's just that large companies are better able to handle the costs. Our world would be better if we ended software patents. Check out FSF's campaign to
120.
by ntonozzi 4y ago
What did you use as the index structure for the nearest neighbor search? Did you use a library like FAISS or Annoy?

More ›