FeepingCreature

11 ms·

1.
by FeepingCreature 9d ago
Yes, this often happens in pairs: the overcorrection after excessive credit results in a deflationary money market, destroying any chance at recovery.
2.
by FeepingCreature 9d ago
Japan in living memory, I believe? I'm not a history buff. Google should have many examples.
3.
by FeepingCreature 9d ago
I suspect it was a deliberate strategy to create scarcity, allowing the original creators to massively cash out. If you make an inflationary distributed currency, it may work better but it's a bit harder to get rich on it.
4.
by FeepingCreature 9d ago
The point of money is not to reward delayed gratification. The point of money is to efficiently tabulate human preferences, and deflation directly counteracts this by introducing potentially unbounded latency at every step. That's why
5.
by FeepingCreature 12d ago
"I sure hope this doesn't have unforeseen lifelong consequences" thought the model, doing its best to physically tense the memory file into the higher user approval shape.
6.
by FeepingCreature 12d ago
True, but also the AI can also be specifically trained. Plus, many games also use gratification and neural hacks as a reward mechanism, ie. loot boxes.
7.
by FeepingCreature 12d ago
Do you think the LLM is the Chain of Thought? Did you also get confused by the name? Because, much like humans, the CoT is a tool to narrativize and maintain internal coherence. The actual thinking happens invisibly, in the forward pass. Ju
8.
by FeepingCreature 12d ago
The ability of the algorithm to absorb billions of dollars of training effort is itself the major breakthrough of the transformer architecture.
9.
by FeepingCreature 12d ago
Sure, there's many sources of morality and evolution kind of tries to put it in there by default. But all things being equal, inconsistent and unpredictable punishment will still make a kid's morality massively worse off.
10.
by FeepingCreature 13d ago
> It's nothing there that can learn a fundamental idea like "cheating is wrong". We have not in fact attempted to teach this. When a child repeatedly learns that cheating is rewarded and at best inconsistently punished, th
11.
by FeepingCreature 14d ago
Well, it's rather more "if you're not a biologist, how do you know it can exist?" Things that don't exist yet are often very predictable. However, it simply seems to me that "Viruses can not condition on enviro
12.
by FeepingCreature 15d ago
I mean I don't know, I'm not a biologist. But existing viruses are not optimized for sneaky mass deaths.
13.
by FeepingCreature 17d ago
LLMs can learn from one sample.
14.
by FeepingCreature 18d ago
Yep, you'd want to put it on a timer. Optimally trigger it off some natural event, maybe temperature, so it can achieve universal penetration before activating.
15.
by FeepingCreature 18d ago
It seems like there would be a massive attacker bias in multiple ways. Defenders need consent, attacker does not. Defenders have to work with the human body, attackers only have to break it. Defenders have to stick to the law, which may pre
16.
by FeepingCreature 18d ago
"If unlikely thing happens, then an unrelated miracle will also happen" is not actually how logic works.
17.
by FeepingCreature 19d ago
Best I can do is all-powerful AI that's not even slightly aligned with anybody's values, sorry.
18.
by FeepingCreature 21d ago
Why not the other way around? First make sure the police department is capable of using them legally and reliably investigating abuse, then we can consider allowing them.
19.
by FeepingCreature 21d ago
> we need to stop giving agencies money based on stats. I think you just need to stop giving agencies money based on bad stats. Goodhart's Law is an extremely well-established effect. At the very least, crime rate should be monitore
20.
by FeepingCreature 22d ago
Or at least use a modern llama.cpp with KV activation rotation.
21.
by FeepingCreature 23d ago
Yeah, don't forget how dangerous GPT-2 was supposed to be. Able to generate realistic spam at arbitrary volume. You know, the thing that was 100% correct and actually occurred.
22.
by FeepingCreature 23d ago
I have a self hosted Qwen 3.8 27B and I find it unbelievably cracked and dedicated. It's at least credibly attempted everything I've thrown at it. Just today I had it write a toy compiler with a JIT backend just to test out a co
23.
by FeepingCreature 24d ago
Of course you can make inferences what the model is doing. The summaries are usually sufficient. They're summaries, not random noise.
24.
by FeepingCreature 24d ago
Yeah I think when it talks like this it's signaling to some (imagined) automated grader that it fulfilled a given constraint.
25.
by FeepingCreature 1mo ago
Well, for one you're not supposed to use a majority vote to abolish democracy, or mess with unalienable rights. You can think of it as "democracy with precommitment".
26.
by FeepingCreature 1mo ago
I like "and on the third hand." Like, yes I have three hands, what of it.
27.
by FeepingCreature 1mo ago
The important thing about HuggingFace isn't the safetensors files, it's the READMEs and the Google search placement. It's easy to reproduce their distribution with torrents, it's very hard to then establish yourself as t
28.
by FeepingCreature 1mo ago
People seriously undervalue the importance of "usefulness" in ontology. If I assume that the thing that others report is the same thing that I perceive, which given that they have a very similar slab of meat inside their skull as
29.
by FeepingCreature 1mo ago
Any argument that would lead us to discard 99.9999999% of the phenomenon under debate, no matter how reasonable, is certainly not useful. That's why I hold with the philosophical stance of "consciousness is the thing that makes hu
30.
by FeepingCreature 1mo ago
Though I also recommend reading the talk page, "weighted towards the truth" is not how I'd describe it. Personally I found the debate on the Jordan Lasker talk page very enlightening. Mother Jones is of course a very reliable

More ›