LessWrong 2.0 Reader

Who is pushing for totalitarianism? I dispute that AI safety people are pushing for totalitarianism.

If you have recommendations, post them! I doubt the author tried to filter the subjects very much by "book subjects" it's just what people seem to have found good ones so far. 

Flippant response: people pushing for human extinction have never been dead under it, either.

I was enthusiastic about the title of this post, hoping for something different from the usual lesswrong content, but disappointed by most of the examples. In my view if you take this idea of learning tacit knowledge with video seriously, it shouldn't affect just how you learn, but what you learn, rather then trying to learn book subjects by watching videos.

In that post, you say that you have a graph of ${10}^{100}$ vertices with a particular structure. In that scenario, where is that structured graph of ${10}^{100}$ vertices coming from? Presumably there's some way you know the graph looks like this

rather than looking like this

If you know that your graph is a nice sparse graph that has lots of symmetries, you can take advantage of those properties to skip redundant parts of the computation (and when each of your ${10}^{100}$ nodes has at most 100 inbound edges and 100 outbound edges, then you only have on the order of a trillion distinct nodes (if we consider e.g. (0, 0, 0, ..., 0, 0, 1) to be identical to (0, 0, 0, ..., 0, 1, 0, ..., 0, 0, 0)).

It's probably worth looking at the process which is generating this graph, and figuring out if we can translate the output of that process directly to a coordinate in our 100-dimensional space without going through the "translate the output to a graph, and then embed that graph" intermediate step.

See also Rational Polarization.

I think charter cities are a questionable idea, even though I'm pro free markets. It seems that the sort of constitional change and stability required for a charter city is no easier to achieve then the kind of constitutional change and stability required for a free market in the entire country. I don't think trying either in developing countries as an outsider is a good use of anyone's resources.

Solidier

Misspelled.

That's actually not (that much of) a crux for me (who also thinks it's mildly manipulative, but, below the threshold where I feel compelled to push hard for changing it).

I'm actually not familiar with the nitty gritty of the LLM forecasting papers. But I'll happily give you some wild guessing :)

My blind guess is that the "obvious" stuff is already done (e.g. calibrating or fine-tuning single-token outputs on predictions about facts after the date of data collection), but not enough people are doing ensembling over different LLMs to improve calibration.

I also expect a lot of people prompting LLMs to give probabilities in natural language, and that clever people are already combining these with fine-tuning or post-hoc calibration. But I'd bet people aren't doing enough work to aggregate answers from lots of prompting methods, and then tuning the aggregation function based on the data.