zahlman

29 ms·

871.
by zahlman 5mo ago
> I don't create new accounts, I never cross-login with my email address ... you don't create burner email addresses specifically to cross-login with them to one service?
872.
by zahlman 5mo ago
What if I only have baking soda around normally, and not baking powder? What if I have bottled lemon juice and can't zest an actual lemon? Also I hoped to see heavy cream listed as an ingredient.
873.
by zahlman 5mo ago
> There's so much click-baity LLM language You're absolutely right. It's not just Claude that gives you output like this — it's every model. (Ahem, sorry.)
874.
by zahlman 5mo ago
> About the only grace I give these manufacturers is the fact that google and apple both make it an annoying pain to maintain applications in their app store. A manufacturer can't simply drop "oven app" once and expect it
875.
by zahlman 5mo ago
The comment you are replying to doesn't argue any such thing, and is pretty clear in its explanation of how your position is perfectly compatible with what is requested.
876.
by zahlman 5mo ago
As a point of reference, 1MLOC is about the size of the entire Python standard library including tests , as well as stuff like IDLE. (Well, the Python part of the code. There's about half that much again of C in Modules/ .)
877.
by zahlman 5mo ago
> VPRI's STEPS project The what now? Search engines failed me here.
878.
by zahlman 5mo ago
I don't see why it would. Did C60? Are fullerenes known for unusual reactivity? I'd expect the opposite, if anything.
879.
by zahlman 5mo ago
It's quite "uniform" (symmetrical). It's essentially the same structure as the 60-atom ball, but with an additional atom at the centre of each hexagon.
880.
by zahlman 5mo ago
It seems to me like they're saying the agent made the tool call they expected, but the harness didn't reject it like they expected it to.
881.
by zahlman 5mo ago
A lot of the time people relate an anecdote about how Claude helped do some cool thing, my reaction is that it's not a thing I would have thought about doing in the first place, and that I still can't really imagine wanting to do
882.
by zahlman 5mo ago
> a welcome farewell to Stack Overflow. Nothing will change the fact that beginners have unknown unknowns. They can't solve most of their problems with a chatbot because they don't know what to ask. Maybe they can literally c
883.
by zahlman 5mo ago
I meant: why tell Claude to edit CLAUDE.md to add the "don't add a signature" rule, instead of just writing a "don't add a signature" rule into CLAUDE.md directly? The necessary rule can't be much longer t
884.
by zahlman 5mo ago
> I'm a little slow so it took me until this week to think of a better version of this: at the end of the project I now ask Claude to write up from scratch a detailed but high-level explanation of what problem we were solving and wh
885.
by zahlman 5mo ago
> However FWIW LLMs are generally good at following a specific style when given examples. In your experience, is it worthwhile to have an agent create a "skill" for itself for following the style? Or is it a better use of conte
886.
by zahlman 5mo ago
> I remember one case where PG&E got approval to charge for grid maintenance, spent hundreds of millions, had ~$100M leftover, then declared "oh we did it cheaper than we expected, executive bonuses all around the C-suite with t
887.
by zahlman 5mo ago
> And until new solar is cheaper than existing coal (which will be awhile, maybe never) then coal only decreases as plants shut down. Why wouldn't "existing solar is cheaper than existing coal, and existing coal is not required
888.
by zahlman 5mo ago
...Do we really need steel just to mount solar panels?
889.
by zahlman 5mo ago
Rather than guessing, have you considered asking ? Sure, sometimes people haven't reasoned themselves into positions, and won't simply be reasoned out of them. But understanding others' epistemology is still the first step t
890.
by zahlman 5mo ago
> Take a look at California. Their grid is routinely, daily, generating ~84% of its power from renewables [1] (with ~25GW of existing solar PV capacity, ~6GW of wind, and ~6GW of hydro). > ... Not everywhere is California, but solar a
891.
by zahlman 5mo ago
The best time to solve personal problems with those techniques, perhaps. That is not the same as (and might be opposed to) the best time to seek employment or go into business using those techniques; there isn't going to be much of a m
892.
by zahlman 5mo ago
> it's not as good as I remember it. They barely touch on the issues of consciousness, Picard basically says "What if Data is conscious?" and then goes off on a tangent. ST:TNG writing is generally like this. The show requ
893.
by zahlman 5mo ago
> Virtue ethics: even if LLMs are not conscious, we should not abuse or mistreat them them because cruelty practised on anything trains one's disposition toward cruelty. If they are not conscious, then how do we know what constitute
894.
by zahlman 5mo ago
> Are you sure that AI-consciousness implies a responsibility to not make them suffer? This is largely the point of describing something as "conscious", yes.
895.
by zahlman 5mo ago
Labeling and categorizing things doesn't change their nature. But the fact that people want to do it is revealing. The only purpose that can really be served by arguing "the LLM system is conscious , you see" is to prop up c
896.
by zahlman 5mo ago
The post speaks of "subjective experience of reality", not "subjective emotion". Both the emotions and the non-emotions listed would fall under that category.
897.
by zahlman 5mo ago
OP wants to assert that it's "important" for these systems to have those qualities, while completely brushing aside the question of whether such systems can in principle actually have those qualities (or their opposites). W
898.
by zahlman 5mo ago
> It's a waste of time to think about whether an LLM has a subjective experience of reality... It's important that Claude is happy, empathic, demonstrates understanding and empathy for the human condition, because we are entrus
899.
by zahlman 5mo ago
We've previously had e.g. https://blog.andymasley.com/p/individual-ai-use-is-not-bad-f... TFA is light on statistics and can only describe the water usage in terms like: > AI data centers require vast amounts o
900.
by zahlman 5mo ago
> I run my home automation network entirely offline, so anything that needs the internet doesn't get added to my cart. Absolutely support you in that. I don't really feel the urge to automate appliances around me in the first p

More ›