I really like learning new things!
Jacob G-W
I’m actually working on this problem right now. There are a lot of those services, but they usually generate bad cards. I’m researching how to use AI to make good cards.
That’s really interesting! Did you ever use Anki or a spaced-repetition app? I wonder if the feeling happens because the brain gets rewarded for having a certain representation? Or did it just appear out of nowhere?
I think the idea actually works pretty well with superintelligence (with one big exception if you assume we all die). Lots of people don’t understand how/why superintelligence could kill us all. They naively think that creating a superintelligence would be a great idea. If we all died, then they would understand why alignment is a necessary complexity. The only problem with this is that we are all dead.
I represent history timelines logarithmically in my head. I talk about it in https://www.lesswrong.com/posts/j8WMRgKSCxqxxKMnj/what-i-think-about-when-i-think-about-history . I think the reason that number lines are logarithmic is because we are more familiar with events/numbers on one side of the line. I’m not sure that number lines are high-dimensional inside people’s heads. At least for me, it is one dimensional and maybe curved a little around my field of vision.
Wow, this is impressive intuition. Do you know what made you think of compressibility first? Or is it just intuition gained through hard work?
Yes, this is what I meant to say, sorry: typo.
Thanks, I did not know this!
I prefer putting everything into one big deck for a few reasons.
It forces you to do all the cards every day. When I had different decks, I would do the “main” deck every day, but then not do the other decks.
I have not read any research on this, but it seems like interspersing the cards could make the cards less likely to anticipate and force you to learn better.
It also just mixes up the monotony and makes it more fun.
There is a prediction market about this that asks the question Are open source models uniquely capable of teaching people how to make 1918 flu?: https://manifold.markets/JeffKaufman/are-open-source-models-uniquely-cap Thanks to @jefftk for creating it.
There’s a bunch. Here’s one: https://manifold.markets/NealShrestha58d3/sam-altman-will-return-to-openai-by
Same!
It should probably say 2023 review instead of 2022 at the top of lesswrong.
Ah, sorry for the confusion. Thanks!
Update, it seems that the video generation capability is just accomplished by feeding still frames of the video into the model, not by any native video generation.
It seems to do something similar to Gato where everything is just serialized into tokens, which is pretty cool
I wonder if they are just doing a standard transformer for everything, or doing some sort of diffusion model for the images inside the model?
Sure, but they only use 16 frames, which doesn’t really seem like it’s “video” to me.
Understanding video input is an important step towards a useful generalist agent. We measure the video understanding capability across several established benchmarks that are held-out from training. These tasks measure whether the model is able to understand and reason over a temporally-related sequence of frames. For each video task, we sample 16 equally-spaced frames from each video clip and feed them to the Gemini models. For the YouTube video datasets (all datasets except NextQA and the Perception test), we evaluate the Gemini models on videos that were still publicly available in the month of November, 2023
Yes, this is pretty much how I see trust. It is an abstraction over how much I would think that the other person will do what I would want them to do.
Trusting someone means that I don’t have to double-check their work and we can work closer and faster together. If I don’t trust someone to do something, I have to spend much more time verifying that the thing that they are doing is correct.
This is super interesting and I have a question:
How difficult would it be to also apply this to the gamates and thus make any potential offspring also have the same enhanced intelligence (but this time it would go into the gene pool instead of just staying in the brain)? Does the scientific establishment think this is ethical? (Also, if you do something like this, you reduce the homogeneity of the gene pool which could make the modified babies very susceptible to some sort of disease. Would it be worth it to give the GMO babies a random subset of the changes to increase variation?)
Thanks for the update! I have a few questions:
In last year’s update, you suspected that alignment was gradually converging towards a paradigm. What do you think is the state of the paradigmatic convergence now?
Also as @Chris_Leong asked, does using sparse autoencoders to find monosemantic neurons help find natural abstractions? Or is that still Choosing The Ontology? What, if not these types of concepts, are you thinking natural abstractions are/will be?
Hey, I’m new here and have really been enjoying reading lots of posts on here. My views have certainly updated on a variety of things!
I’ve been exploring using Anki flashcards to codify my thought processes when I have a-ha moments. After reading about cached thoughts, I started thinking that most of executing procedural knowledge is just having lots of cached thoughts about what to do next. I understand that this is not exactly the type of cached thought in the post, but I think it is interesting nonetheless. I have been making Anki cards like
Physics: what should you do if you get something as a function of x instead of a function of t to solve a problem//use conservation of energy instead
to speed up the process of learning new procedures (like solving physics problems).Have others done something similar?