LessWrong 2.0 Reader

or there is already a datacenter there but it doesn't consume that much energy for now

Seem that way:

AWS literally bought a 1000MW nuclear-powered datacenter campus for $650M USD. Though only the very first building with 48MW of capacity is likely to be online in the near term, this provides a valuable pipeline of datacenter capacity for AWS without being held up waiting for power generation or grid transmission capacity. We think a campus of such mammoth proportions will take many years to fully ramp to the promised 1,000 MW of Critical IT Power.

Assuming we require a performance of 40 tokens/s, the training cluster can run $\frac{2000}{30}\times 24000=1,600,000$ concurrent instances of the resulting 70B model

Nit: you mixed up 30 and 40 here (should both be 30 or both be 40).

I will assume that the above ratios hold for an AGI level model.

If you train a model with 10x as many parameters, but use the same training data, then it will cost 10x as much to train and 10x as much to operate, so the ratios will hold.

In practice, I believe it is universal to use more training data when training larger models? Implying that the ratio would actually increase (which further supports your thesis).

On the other hand, the world already contains over 8 billion human intelligences. So I think you are assuming that a few million AGIs, possibly running at several times human speed (and able to work 24/7, exchange information electronically, etc.), will be able to significantly "outcompete" (in some fashion) 8 billion humans? This seems worth further exploration / justification.

In the late 1940s and early 1950s nuclear weapons did not provide an overwhelming advantage against conventional forces. Being able to drop dozens of ~kiloton range fission bombs in eastern European battlefields would have been devastating but not enough by itself to win a war. Only when you got to hundreds of silo launched ICBMs with hydrogen bombs could you have gotten a true decisive strategic advantage

Amazon recently bought a 960MW nuclear-powered datacenter. 

I think this doesn't contradict your claim that "The largest seems to consume 150 MW" because the 960MW datacenter hasn't been built (or there is already a datacenter there but it doesn't consume that much energy for now)?

Great!

I am contrasting generating an output by:

1. Modeling how a human would respond (“human modeling in output generation”)
2. Modeling what the ground-truth answer is

Eg. for common misconceptions, maybe most humans would hold a certain misconception (like that South America is west of Florida), but we want the LLM to realize that we want it to actually say how things are (given it likely does represent this fact somewhere)

Adopted.

Nevertheless lots of people were hassled. That has real costs, both to them and to you. 

i like the idea. it looks useful and it fits my reading style well. i wish something like this were more common - i have seen it on personal blogs before like carado's.

i would use [Concept Dependency] or [Concept Reference] instead so the reader understands just from seeing the title on the front page. also avoids acronym collision

If that were true then there are many ways you could partially do that - eg give people a set of tokens to represent their mana at the time of the devluation and if at future point you raise. you could give them 10x those tokens back.