🤓🔫 "exactly 499" according to google docs which seems to count contractions as single words.. basically split by space but e.g. is 2 words as far as I can tell, but if I and have were to be counted as 2 words, then the precommitment gives John a tiny little fuck right back:
the famous case where a guy lost all his emotions and in many ways things remained fine.
I believe the point of the example was that the guy would take forever to make any decision, even small things like what to pick in a store, wasting hours on trivial stuff (..and when I imagine my lack of patience with friends when they are indecisive sometimes, I suspect his social life might not "remained fine" but that was not mentioned in the interview),, that emotions (like value functions in ML) help to speed up progress even if with unlimited time we might get to the same conclusions
Funny perspective,, but why not model the phenomenon as distributed computing instead of using loaded labels? Mundane things like what to eat for dinner is a collective family-unit decision, modelling is as bi-directional dominance game sounds counter-useful to me, even though the description could be made equally accurate in both frames.
Say, when I cook pasta an hour sooner than I would have otherwise when my husband is hungry already, I even delegate 99% of the decision about various health hazards to the institutions of the civilization I live in - the tap water is fine, the bag of pasta is fine, the tomatoes are meh but fine, I just check the non-moldy cheese is indeed not moldy yet. I don't have a chemical and bio lab to check for all the poisons and microbes myself, yet I don't see it as being submissive to the society, I just trust the external computation that made the decisions for me is aligned enough with my interests..
In the absence of other perspectives on downsides, I would like to mention that blunt memes that are catchy phrases can lead to polarization.
Perhaps better "ammunition" would be silent memes that are building blogs of working institutions - when I buy a loaf of bread, there is no catchy phrase "buy our bread, it contains no anthrax" said by anyone anywhere anytime ever... yet the silent implication is true, I will, in fact, not get any anthrax with my bread. And the bigger picture implied by that silly example is an egregore of the boring institutions of the civilization that I rely upon for my own safety every day, the existence of which implies the existence of memeplexes that encode for it, but there is no implication of memes in the form of catchy English phrases.
It might well be the case that a fight of catchy phrases is a game created by a memeplex that favours successionism phenotype - what if LLMs are better at generating words and images than building and maintaining humane institutions..?
1Would it useful to think about (pre-trained) LLMs as approximating wave function collapse algorithm? (the one from game dev, not quantum stuff)
Logits as partially solved constraints after finite compute budget and output is mostly-random-but-weighted-towards-most-likely sample without actually collapsing it fully and without backtracking and each node is evaluated to random level of precision - basically a somewhat stupid way how to sample from that data structure if you don't follow it by fixing the violated constraints and only keep the first pass of a quick heuristic, there will be incompatible nodes next to each other... as in hallucinations and harmful mixing of programming paradigms in the same codebase and 80%-good-enough stuff that could not possibly be precise in edge cases.
And stuff like RLHF or RLVR will still only improve the first pass heuristic, not actually fix the inconsistencies ... "agentic" scaffolds for coding assistants with multiple passes and running the linters and tests and multiple rounds of "does it make sense" sound like they should be helpful, but doing it in tokens instead of logits (where the actual contraints live before collapsing them to quasi-random instantiated sample) sounds ..inefficient?
Dumping on normies isn't particularly interesting most of the time,
I thought gossip was the thing for the human brain - personally, I find these ways of digging your own grave quite fascinating 🍿
fantasies live in a separate magisterium where they suspend disbelief, and they never think about how to actualize those fantasies
to overthink potential reasons for the observed ratio of "fantasy -> planning" and "separate magisteria" - I imagine that the genes that correlated with agency about fire fantasies found themselves in burning villages..
reductio ad absurdum for the idea of wanting more agency about fantasies - "how to make a thinking machine?" .. perhaps we need more grandmothers to slap bright-eyed boys on the wrists, not to pour more billions of dollars into agentic fantasies?
but I agree there is enough room for benign fantasies to make it into planning (..but then perhaps keep the plan under a lid for most stuff, like for a fantasy that involves stripping John off his sunglasses [not you as a person, John as a para-social sexualized object] - the appeal is only in the imagination, not that I would want to attempt doing it for real)
1🌶️take inspired by https://www.anthropic.com/engineering/code-execution-with-mcp
agents should (be trained to actually) RTFM before touching existing code (i.e. do the equivalent of mouse hover for signatures and docs of just-faded-from-memory functions) instead of vibing the type from far away context (or let's be honest, just guessing from function name and the code so far and being good at guessing)
I hope the next fashion wave will go for short short-term memory while really "using tools" instead of long short-term memory with "yolo tools as special tokens never seen in pre-training"
I don’t think most peoples’ thinking most of the time routes through the (fantasy) -> (planning) move.
relative thinking time does not sound like a useful measure - if I imagine someone made a real living dragon, would I want to ask how many hours did they spend on thinking about "How would I make a real living dragon?" compared to thinking about "the rest of their life" .. meh 🤷
also if I consider 2 workflow variants:
- 16yo: (fantasy) -> (planning) -> (learning it's not possible the hard way)
- 40yo: (fantasy) -> (instantly notice the weakest part of potential planning)
..is it "better" to spend time on pre-doomed projects by not knowing better yet or to miss learning opportunities by focusing only on feasible projects?
..maybe you still like "time" measure here, but do you also have a better measure in mind how to stay on a healthy trajectory between optimism/curiosity and grounding/focus?
why focus only on the brains? it's a property of the mind and I thought the standard take why humans are not even approximating utility maximizers is because of properties of the environment (priors), not a hardcoded function/software/architecture/wetwere in the brain .. or?