All of Trevor Cappallo's Comments + Replies

People aren’t going to read books and stop to ask questions. That sounds like work and being curious and paying attention, and people don’t even read books when not doing any of those things.

People definitely aren’t going to start cracking open history books. I mean, ‘cmon.

The ‘ask LLMs lots of questions while reading’ tactic is of course correct.

 

I was thinking along these lines about a year back, and I started working on an ePub PWA (web-based) reader with some bells and whistles. The relevant whistle here is that you can highlight a word, passage, ... (read more)

I didn't find the results about cheating and shoplifting surprising, but that tracks with my friend group at the time. That said, I was curious about whether there's a gender discrepancy in shoplifting (there's not), and found a large 2002 survey which gives 11% as the lifetime incidence of shoplifting in the U.S.

I confess I am perplexed, as I suspect most people are aware there is more than one Trevor in the world. As you point out, that is not your last name. I have no idea who you are, or why you feel this is some targeted "weaponization."

Is it conceivable that this is purely an emergent feature from LLMs, or does this necessarily mean there's some other stuff going on with Sydney? I don't see how it could be the former, but I'm not an expert.

Long before we get to the “LLMs are showing a number of abilities that we don't really understand the origins of” part (which I think is the most likely here), a number of basic patterns in chess show up in the transcript semi-directly depending on the tokenization. The full set of available board coordinates is also countable and on the small side. Enough games and it would be possible to observe that “. N?3” and “. N?5” can come in sequence but the second one has some prerequisites (I'm using the dot here to point out that there's adjacent text cues show... (read more)

My best guess is that there's a metaverse which consists of (at a minimum) every possible computation. While not technically provable or falsifiable, it does result in predictions which mean that circumstantially we should have an excellent guess whether or not it's true.

So far, it's true. It nicely explains the fine-tuned constants and QM and the discrete nature of the apparent finest (Planck-region) levels of reality. And yes, it also predicts that we will, on average, be overwhelmingly likely to live in one of the simplest possible universes supporting ... (read more)

From what I know of security, any system requiring secrecy is already implicitly flawed.

(Naturally, if this doesn't apply and you backchanneled your idea for some legitimate meta-reason, I withdraw my objection.)

1benjamincosman
I think secrecy is rarely a long-term solution because it's fragile, but it can definitely have short-term uses? For example, I'm sure that some insights into AI have the capacity to advance both alignment and capabilities; if you have such an insight then you might want to share it secretly with alignment researchers while avoiding sharing it publicly because you'd rather Facebook AI not enhance its capabilities. And so the secrecy doesn't have to be a permanent load-bearing part of a system; instead it's just that every day the secrecy holds up is one more day you get to pull ahead of Facebook.

For the record, I found that line especially effective. I stopped, reread it, stopped again, had to think it through for a minute, and then found satisfaction with understanding.