leetharris

8 ms·

61.
by leetharris 2y ago
> Plus, how exactly did Deepseek lie. The model size, data size are all known. Calculating the number of FLOPS is an exercise in arithmetics, which is perhaps the secret Deepseek has because it seemingly eludes people. Model parameter co
62.
by leetharris 2y ago
If we're going to play that card, couldn't we also use the "Chinese CEO has every reason to lie and say they did something 100x more efficient than the Americans" card? I'm not even saying they did it maliciously, b
63.
by leetharris 2y ago
Could be true. Deepseek obviously trained on OpenAI outputs, which were originally RLHF'd. It may seem that we've got all the human feedback necessary to move forward and now we can infinitely distil + generate new synthetic data
64.
by leetharris 2y ago
CEO of Scale said Deepseek is lying and actually has a 50k GPU cluster. He said they lied in the paper because technically they aren't supposed to have them due to export laws. I feel like this is very likely. They obvious did some gre
65.
by leetharris 2y ago
Laravel has something like this called Livewire. It's excellent. Laravel is so much better than Django, but I just can't go back to PHP at this point. https://livewire.laravel.com/
66.
by leetharris 2y ago
I see what you're saying, but I also feel like ALL Nvidia cards are "2K" oriented cards because of DLSS, frame gen, etc. Resolution is less important now in general thanks to their upscaling tech.
67.
by leetharris 2y ago
Generally yes. That being said, sometimes multimodal LLMs show decreased performance with extra modalities. The extra dimensions of analysis cause increased hallucination at times. So maybe it solves the frog problem, but now it's hall
68.
by leetharris 2y ago
I think it really depends on whether or not you can offer a competitive solution and what your end goals are. Do you want an indie hacker business, do you want a lifestyle business, do you want a big exit, do you want to go public, etc? It
69.
by leetharris 2y ago
The Google ASR is one of the worst on the internet. We run benchmarks of the entire industry regularly and the only hyperscaler with a good ASR is Azure. They acquired Nuance for $20b a while ago and they have a solid lead in the cloud spac
70.
by leetharris 2y ago
A few problems with this approach: 1. It brings everything back to the "average." Any outliers get discarded. For example, someone who is a circus performer plays fetch with their frog. An LLM would think this is an obvious error
71.
by leetharris 2y ago
The main challenge with using LLMs pretrained on internet text for transcript correction is that you reduce verbatimicity due to the nature of an LLM wanting to format every transcript as internet text. Talking has a lot of nuances to it. J
72.
by leetharris 2y ago
People are more than welcome to voluntarily stop eating meat! I'm curious, do you eat meat? I was vegetarian for a while and didn't mind it.
73.
by leetharris 2y ago
There's a lot of Mac games on Steam, Apple Arcade, and Battle.net these days. Anything that isn't supported there, I generally use Xbox streaming or GeForce Now streaming. Here's a list of my most played games on my Mac in th
74.
by leetharris 2y ago
Posted in another comment: These Cable Matters ones work well, but need their firmware flashed. Luckily, it was extremely easy and worked great for me. Here's an Apple forum thread about it: https://forums.macrumors.com&#x2F
75.
by leetharris 2y ago
These Cable Matters ones work well, but need their firmware flashed. Luckily, it was extremely easy and worked great for me. Here's an Apple forum thread about it: https://forums.macrumors.com/threads/dp-usb-c-thun
76.
by leetharris 2y ago
It's not a dumb question. I actually used to use an iMac 27" with an Nvidia 680 that I would boot into Bootcamp / Windows for my primary gaming computer. I covered it in "built, not bought" stickers at Quakecon one
77.
by leetharris 2y ago
I have a Steam Deck, Asus ROG Ally, M2 Mac Mini, M1 Pro laptop, M2 Max laptop (work). All of this runs on either an LG C3 42" OLED or a 34" 1440p ultrawide. Linux GeForce Now can only do 720p or 1080p, can't remember which. A
78.
by leetharris 2y ago
Love to see that it still starts at $599. My M2 Mac Mini that I got for $499 is my favorite gaming computer I've had in a long time. Runs many games like WoW, Dota, League of Legends, etc great. Anything that it doesn't run due to
79.
by leetharris 2y ago
Yes I've seen it a couple of times and it is unbelievably expensive and always a mess. The amount of Salesforce consultants, Salesforce developers, plugins, tools, etc to keep it all stable is wild. Still, there's a reason people
80.
by leetharris 2y ago
I am very biased as I am from the Austin area, but this is a big reason why I still live here. Everywhere here seems to be very hybrid friendly and the suburbs are close and very cheap in comparison to those 3 cities.
81.
by leetharris 2y ago
People keep making these, but they become outdated so fast and nobody keeps up with it. If your definition of "great" changes in 6 months because a new model shatters your perception of "great," it's hard to rescore
82.
by leetharris 2y ago
Yeah the 3B are NOT open. The 8B are as they can be used under commercial license.
83.
by leetharris 2y ago
In general I feel like all model providers eventually become infrastructure providers. If the difference between models is very small, it will be about who can serve it reliably, with the most features, with the most security, at the lowest
84.
by leetharris 2y ago
> Was there a similar backlash against the rise of Java, Swift, Kotlin, or TypeScript? Absolutely. Swift was hated because it lacked so much Objective-C functionality at launch, Kotlin got a similar "Java is fine" reaction, and
85.
by leetharris 2y ago
We released a new SOTA ASR as open source just a couple of weeks ago. https://www.rev.com/blog/speech-to-text-technology/introduci... Take a look. We'll be open sourcing more models very soon!
86.
by leetharris 2y ago
Hmm, I understand what you mean. But I think there's a difference between complexity and optionality / versatility. For example, it has different ways to declare functions because assignment is generally consistent (and IMO easy t
87.
by leetharris 2y ago
Now that I'm thinking about it, most of it is probably .NET bloat instead of C# bloat, but a few examples would be global usings, file scoped namespaces, records, target-typed new expressions, null coalesce assignments, etc. It's
88.
by leetharris 2y ago
> Everybody who does Express, React, or any other popular advanced libraries with TypeScript is using these features. This is very true and my original post was short sighted. You could, of course, make most upstream dependencies without
89.
by leetharris 2y ago
This is easily the most appealing thing to me about Go. I learned Go through the "Learn Go with Tests" way and I had a ton of fun. It is hard for me to recommend using Go internally since .NET/Java are just as performant and
90.
by leetharris 2y ago
> I'm not so sure about that. I think we end up consuming a lot of these features in the TS types that get published alongside libraries. We just don't know it, we just get surprisingly intuitive type interfaces. Very true. As

More ›