They use a lightweight adapter to silently degrade the performance. Usually these adaptors are made to improve the performance for a given domain/task.
It might be possible to train a big generalist that is a composition of modules, some of which can be dropped dynamically at inference time, depending on the prompt.
Cool. Thanks for sharing. I am thinking about creating a series of smaller models for specific purposes and then orchestrating them so they mirror the human brain which is a bunch of subsystems that give multiple opinions about the same stimulus
I agree with the intent of your rhetorical question, so I'm jesting with you. I'm justifying my "yes" with the hopefully humorous distraction that every person, including American taxpayers, has at some point made a nonsustainable/selfish (my definition of immoral) decision.
My big gripe with unions is the unwavering protection of their worst performing members.
Eg, that they necessitated so called "rubber rooms" like these in the NYC public schools, where teachers got paid to do nothing while waiting on arbitration.
I doubt you'll find many people in favor of how bad cops get protected by police unions either. At least in the US I'd much rather a broad social net so my health care and retirement weren't so directly tied to my job than a union specific to my trade.
It’s not really about this particular claim. It’s that I can read a comment that has a reasonable chain of logic and I don’t know if it’s true. This topic is just not easily studied and theories are hard to falsify.
In all fairness most of the unique stuff I can do is probably an artifact of my training process, so it seems unfair to deny an LLM the same accomodation.
This got me thinking, and it might actually even be a comparable amount.
Let's estimate 12 years of schooling run at minimum $100,000 per student, at least in the US [1], and then add onto that number whatever else you may do after that, i.e. a bunch more money if paid (college) or "unpaid" (self-taught skills and improvements) education, and then the likely biggest portion for white-collar workers, yet hard-to-quantify, in experience and "value" professional work will equip one with.
Now divide the average SOTA LLM's training cost (or a guess, since these numbers aren't always published as far as I'm aware) by the number of users, or if you wanted to be more strict, the number of people it's proven to be useful for (what else would training be for), and it might not be so far off anymore?
Of course, whether it makes sense to divide and spread out the LLMs' costs across users in order to calculate an "average utility" is debatable.
In the last step of training LLMs, reinforcement learning from verified rewards, LLMs are trained to maximize the probability of solving problems using their own output, depending on a reward signal akin to winning in Go. It's not just imitating human written text.
Fwiw, I agree that world models and some kind of learning from interacting with physical reality, rather than massive amounts of digitized gym environments is likely necessary for a breakthrough for AGI.
reply