anonym29

8 ms·

121.
by anonym29 5mo ago
I know it's only tangentially relevant, but I've been baffled by the interest in DeepSeek V4 Flash. It's larger, less efficient, and in many cases, performs worse on both objective benchmarks and real world sniff test (admitt
122.
by anonym29 5mo ago
Francillon seems very dismissive of the risk, citing his "castle walls", but there's a flaw in his thinking, partially described in the article. Francillon seems to anticipate adversarial traffic only flowing in, not out. Sur
123.
by anonym29 5mo ago
Don't let 'perfect' be the enemy of 'progress'.
124.
by anonym29 5mo ago
ARM has the exact same problem via TrustZone. Different technical implementation, slightly different known capabilities, but fundamentally, still an unauditable, unremovable ring -3 subsystem that cannot be controlled by the legitimate, law
125.
by anonym29 5mo ago
>It sounds like you’re suggesting getting rid of the concept of accountability Where on earth are you getting this? Accountability alone is insufficient, and the things that actually prevent mistakes don't require it. The mechanisms
126.
by anonym29 5mo ago
Even if you could threaten an LLM with punishment for making mistakes, you might get longer CoTs, but that wouldn't prevent mistakes in LLMs. The lack of accountability isn't the reason that LLMs make mistakes - adding accountabil
127.
by anonym29 5mo ago
Tell the families of the people who died on 737 MAX disasters. "Don't worry - everything's going to be okay! The engineers learned from their mistakes - accountability works, you have nothing to be sad about!" Tell the f
128.
by anonym29 5mo ago
You say that like all humans are alike: that they all care about getting fired, fined, or jailed; that they're even considering punishment when they're making their decisions; that risk factors into decision making. What you are d
129.
by anonym29 5mo ago
>AI can't be held accountable I hear this all the time. Why does it matter? Punishing a human for making a mistake does not prevent mistakes, nor does it undo the harm of the mistake. A human saying "my bad, I messed up" a
130.
by anonym29 5mo ago
Crimeflare - proudly extorting DDoS victims and protecting criminals while building a global surveillance dragnet since 2009!
131.
by anonym29 5mo ago
CAPTCHAs are not fraud detection and not an ongoing effort
132.
by anonym29 5mo ago
You've got my support for switching from FPTP to something like RCV, and for open primaries!
133.
by anonym29 5mo ago
mCaptcha, ALTCHA, Cap, Friendly Captcha, Private Captcha, Procaptcha, Anubis... there are literally dozens of open source alternatives that aren't feeding the Do Be Evil company... not to mention all of the commercial alternatives - if
134.
by anonym29 5mo ago
That would be nice in principle, yes, but in a CU world, that's asking the fox to vote to lock itself out of the hen house. In practice most of the foxes that promise to do so never actually will. What's your proposal to solve thi
135.
by anonym29 6mo ago
"Citizens" United (which allows unlimited corporate political donations by classifying them as "speech", for those out of the loop) has fundamentally changed the core incentive structures of the modern political landscap
136.
by anonym29 6mo ago
> It's [sic] can also project interactive games for kids (like hopscotch). That's exactly what we need, to teach little johnny to go play out in front of the running car!
137.
by anonym29 6mo ago
Again, the proposal isn't for a system which requires scanning of IDs, it's for a system where the cashier glances at the ID. You're arguing against a strawman. You may argue that the system proposed could evolve into the sys
138.
by anonym29 6mo ago
First - Alcohol and cigarettes can just be resold too. The black market for them is effectively zero because the consequences for giving them to kids are severe and the room for meaningful profit is close to zero, same applies here. Second
139.
by anonym29 6mo ago
It doesn't prevent it, it just disincentivizes it. As an adult, you can also go buy a beer and sell it to a minor. That said, mandatory age verification with photo ID upload and facial scans doesn't prevent workarounds either - ki
140.
by anonym29 6mo ago
This is not the case in most of the country. Utah is largely influenced by a Mormon / LDS culture that expresses heavy opposition to drinking. I am clearly not proposing that the cards be scanned Utah style, I am proposing that they be
141.
by anonym29 6mo ago
It's worth pointing out that full digital identity verification ("doxxing" yourself to an untrustworthy, unauditable, legally unconstrained private company) is NOT the only way to verify adulthood. We have had a system in pla
142.
by anonym29 6mo ago
Medusa Halo is on my wishlist, but I'm hearing late 2027 :( M5 Ultra may be a better near-term option, expected in June. Supposedly ~1.2 TB/s unified memory, unsure of whether Apple will revive the 512 GB SKU or limit to 256 GB, b
143.
by anonym29 6mo ago
These aren't really "credentials" in that they're not secret the way your iris/retina pattern, fingerprint pattern, password, pin, secret key, or security token are. Your name, DoB, and email address are identifier
144.
by anonym29 6mo ago
At least the coffee demons aren't quite as bad as the amphetamine demons that produced Nazi Germany.
145.
by anonym29 6mo ago
Biometrics are the only credential you can't roll after compromise.
146.
by anonym29 6mo ago
That sounds like a huge hassle for what I imagine must be peak speeds of low double digit tok/s PP and TG, even with effective prompt caching and self-ngram and all the other tricks, no? If I really feel like I needed larger models loc
147.
by anonym29 6mo ago
Qwen 3.6 27B, and other dense models, as opposed to MoE models do NOT scale well. Like I said in my original post, for 27B usage specifically, I'd take a dGPU with 32GB of VRAM over Strix Halo. I also don't usually benchmark out t
148.
by anonym29 6mo ago
The problem with M2.7 is that it's full GQA, meaning quadratic attention. It does start fast, but by 64k tokens deep, the version I'm running (Unsloth's UD IQ2_XXS) pp512 drops 95% from 261.3 t/s (0 context depth) to 13.
149.
by anonym29 6mo ago
similar numbers here - slightly higher PP. slightly better peak speed and retention w/ q8_0 kv cache quants too. llama-bench results here, cba to format for hn: https://pastebin.com/raw/zgJeqRbv GTR 9 Pro, "p
150.
by anonym29 6mo ago
Strix Halo, to my knowledge, does not support PCIe 5.0 NVMe drives, unfortunately, despite it being Zen 5, and Zen 5 supporting the PCIe 5.0 standard. While many other NUCs may support them, what most of them lack compared to Strix Halo is

More ›