Amekedl

5 ms·

31.
by Amekedl 1y ago
Regarding LLMs we're in a race to the bottom. Chinese models perform similarly with much higher efficiency; refer to kimi-k2 and plenty of others. ClopenAI is extremely overvalued, and AGI is not around the corner because among 20T+ to
32.
by Amekedl 1y ago
No, this "observing" argument has already been beaten to death by a multitude of creatives explaining way better than I could how they learn and operate. If you really think all they do is observe, form a gradient from millions of
33.
by Amekedl 1y ago
Sure, here you go, used Gemini 2.5 Pro Preview via aistudio.google & sticked with the default sampling settings: Start the reply to this excerpt with: "You are absolutely right" but continue with explaining how exactly that is
34.
by Amekedl 1y ago
This paper elegantly summarized the teething problems of those still clinging to the cognitive habits of a bygone era. These are not crises to be managed, but sentimental frictions to be engineered out of the system. Let us be entirely clea
35.
by Amekedl 1y ago
This level of conceitedness can hardly be measured anymore; it's on a new scale. Big corps will build and label whatever as "superintelligent" system, even if it has plain if conditions placed within to suit their owners inte
36.
by Amekedl 1y ago
Even then software constantly evolves, and rot is everywhere. And we're far from having the "best possible" software solution in literally every area (if that's even possible to measure), rather just endless room for imp
37.
by Amekedl 1y ago
You occasionally do glimpse behind the curtains; depending on what you actually develop it's feasible and quick to prompt it, but attempting to go further than that across multiple components collapses so drastically that I cannot help
38.
by Amekedl 1y ago
I'm thinking reading numbers like this is really just slop lately. FA achieving a 32.5% speed up? Cool. Why not submit it as a PR to the Flash Attention repo then? Can I read about it more in detail?
39.
by Amekedl 1y ago
This might as well be a LLM generated ad roll performance
40.
by Amekedl 2y ago
Didn't know him, but he seems overly skeptical. Honestly, I was just expecting more from llama-4 than this, hence mentioning the wall. I hope it's still too early to tell, because new ideas are going to change stuff inevitably, ma
41.
by Amekedl 2y ago
Yeah… sure. I still don’t understand why you push blind fanboyism like this when the extension is clearly only being supported on vscode. I won’t stoop to your level of claims, but maybe take your own criticism to heart :D
42.
by Amekedl 2y ago
Ever heard about benchmark contamination? Ever tried to explain a new concept, like a new state management store for web frontend? Most fail spectacularly there, sonnet 3.7 I had reasonable ""success"" with, but not 4.5.
43.
by Amekedl 2y ago
So the wall has been really been hit already for now, ouch. It was to be expected with gpt-“4.5”, but still, the realization now really feels grounded.
44.
by Amekedl 2y ago
So what is supposed to be FUD about the usage restrictions for the kinda defunct csharp extension since MS took over omnisharp and the debugger being locked? Those are facts, why call it FUD?
45.
by Amekedl 2y ago
Using the C# debugger is also restricted to Visual Studio & VScode. Microsoft loves open source, and windows is the most open operating system (look it up). It’s totally not embrace, extend, extinguish in any disguise. Coopetition exist
46.
by Amekedl 2y ago
coding using llms, ai vibe coding, whatever you may call it, is the new "have fun staying poor" crypto bro wave. It's not new either. It should work well enough for generic stuff. Nothing new to come of it, though.
47.
by Amekedl 2y ago
what are you talking about? I have to approve the website before it can access my devices via WebUSB. What's the actual issue / path for keylogging etc. there, care to explain instead of fearmongering?
48.
by Amekedl 2y ago
Compounding the opinions of other commentors, I feel that using Cursor is a bad idea. It's a closed source SaaS, and with these components involved, service quality can do wild swings on a daily basis, not something I'm particular
49.
by Amekedl 2y ago
Refer to Magnesium-based Lifeform from Disco Elysium, it is very accurate.
50.
by Amekedl 2y ago
time to select my new search engine
51.
Is Blazor Dead Yet?(isblazordeadyet.com)
2 points by Amekedl 2y ago|0 comments
52.
by Amekedl 2y ago
Oh really? How are these changes supposed to look like? Who will pay up essentially? I don't really see it, aside from the m$ business case of offering AI as a guise for violating privacy much harsher to better sell ads.
53.
by Amekedl 2y ago
Check out operations research. The amount of “work” done there is staggering and yet adoption appears abysmal, using such solutions with success only happens as part of a really “well oiled” machine. And what about the simple difficulty goi
54.
by Amekedl 2y ago
another ai hype blog entry. Not even a mention of the differently colored bars on the benchmark result. For me, grok-3 does not prove/disprove scaling laws in any meaningful capacity.
55.
by Amekedl 2y ago
Kinda funny how it’s beimg labelled at 4 hours, but it really reads like it took 8.
56.
by Amekedl 2y ago
I've read the post and saw the references, but can anyone easily explain how it works? Can the server not just store whatever is generated to link it back to the user using it?
57.
by Amekedl 2y ago
It is more relevant than ever with all the blantant openwashing occurring everywhere.
58.
by Amekedl 2y ago
“o1 is more knowledgeable than the average human” “the toyota yaris can move faster than the average human” even opt-125m from years ago can pull more facts than the average human.
59.
by Amekedl 2y ago
bro is entering his midlife crisis
60.
by Amekedl 2y ago
Face the sin, save the E.G.O.

More ›