language models

“ULM fit is the wrong approach and that's why we're seeing a lot of these You know, so-called alignment tax, and this view of like, oh, a model can't both code and do other things. You know, I think it's actually because people are training them wrong.”

“I thought, okay, so if I do this at a much bigger scale, using all of Wikipedia, what would it need to be able to do to finish a sentence in Wikipedia effectively, to do it quite accurately, quite often? I thought, geez, it would actually have to know a lot ab…”

“To me, the right way to do this is to fine, fine-tune language models, is to actually throw away the idea of fine-tuning. There's no such thing. There's only continued pre-training.”

“You could use like a hundred times smaller language model and do much better at filtering than RLHF”

“People tried that with language models, which is if you have a prompt and a completion and you just have someone rate it from zero to 10, could you then train a reward model on all of these completions and zero to 10 ratings and see if you could actually chang…”

“It's almost more natural to a CLI than it is in a graphical user interface, because it feels like there's back and forth with the computer. Yeah. Almost funnily like a language model. So I think there's some interesting intersection of like CLIs and language m…”

“We found it to be pretty calibrated. There varies on the model.”

“One of the things we found out about these language models is that not only can you define nodes, it's really good at figuring out what are nodes and what are edges.”

“I mean, the biggest one really was the fact that, like, I think for just four years I was so bearish on language models. And just NLP in general, I was just like, ah, like, none of this really works. Like, why would I spend time focusing on this? I gotta go do…”

“That's like one of the things that's magic about WebSim generally is that it gives language models much far greater tools for expression, right?”

“I think that it is underappreciated how powerful, there's the generative capabilities of language models, but there's also the ability to program them to function as arbitrary machine learning systems, basically for marginally zero cost.”

“We do often normally, in fact, see a 10 X variation in P-ninety latency over the course of half an hour. When we're prompting these models, which is way higher than if you're working with a, you know, a more, more kind of conventional conventionally backed API…”

“This emergent behavior that you're seeing, Is not really emergent behavior, but is really a function of the evaluation metrics that we're using.”

“A difference to language models is probably the amount of context that you need is significantly less than maintaining a long multi-term conversation. And so, you know, coupling this short-term spatial memory with this, like, longer-term object pointers we fou…”

“When you're using these models, if you're getting the answer you want, always, it means you're not asking them hard enough questions.”

“My research says is entirely on this. Like you probably shouldn't trust these models to do the things in adversarial situations.”

“Whenever someone says X is true about language models, you should always append the suffix for current models, because I'll be the first to admit I was one of the people who was very much on the opinion that these language models are fun toys and are going to …”

“I feel like most people who write about language models being good or bad, some underlying message of like, you know, they have their camp and their camp is like, AI is bad or AI is good or whatever. And they like, they spin whatever they're gonna say accordin…”

“So we have seen with the paper called sync attentions that actually the language model allocates a lot of a lot of, because when you do the attention mechanism, you are doing a weighted sum over the tokens, and each token is given a weight, and we see that mos…”

“I feel like in some sense, I feel like prompt engineering, even it's like a slightly negative word at the time, because it refers to all those kind of weird tricks that you have to apply. But I think we don't have to do that anymore. Like given today's progres…”

“Language models frequently will produce like a smaller diff when possible, rather than trying to do a big refactor.”

“Language models have no ability to do this. They are. Kind of per token computation devices where each token is outputted after doing this forward pass and within that there's no explicit structure to hold these intermediate states.”

“The things that language models understand best tend to be the kind of piece of code that are represented on the internet. You know, maybe I don't think a lot of people would be happy with this, that maybe the kind of fundamental language of AI becomes Python …”

“A language model, for example, has no prior for a mouse movement. It, you know, it really never seen that on the internet, but it has a really strong prior for code. So out of the two categories, say web browsing and coding, coding is the only one that is actu…”

“In the era of language models, I don't think you'll be able to train superintelligence from scratch.”

“Over the coming years, there'll be more kind of AI friendly or language model friendly UIs. And what's friendly to a language model is, is code. So the way a model will be doing work, not just for coding and software engineering, Is by basically making functio…”

“So the states are, like, the text prefixes, so the initial states, like, the prompt the actions are the next tokens that means, like, a language model is, like, exactly what a policy is, right? A policy maps a state to a probability distribution of our next ac…”

“Language models are next token predictors is like a fact. Like that is what they do. They are trained to predict the next token. However, that does not mean that they myopically only consider the next token When they choose the next token, you can work on brea…”

“That means that like language models pack a lot more in less space than Vision models. So maybe like a kind of like really hand wavy analogy, right? It's like, well, if you want curve detectors, like you don't need that many curve detectors. You know, if each …”

“If you look at just the neurons of a lot of vision models, you can See neurons that are curve detectors or that are edge detectors or that are high, low frequency detectors. And so you can sort of like make sense of the neurons mostly. But if you look at neuro…”

“If we make this playable, you know, then it kind of can teach people how to use AI, like language models just by playing. Cause you'll like understand how they work. You have to negotiate against them. You see their responses.”

“If you're looking at the context that's being sent to the language model, could you play the game?”

“I don't think we have a lot of evidence that we can build a system like this that like is really, really good at reasoning, but really dumb about the world. Like, I don't know if we have the tools.”

“Like, no matter how you scale the training size, you hit this like perfect, perfect ish plateau in auto memorization, which we call the model capacity.”

“In terms of economic value, programming is definitely the highest cost of benefit for language models right now.”

“So it's like, how do you come up with a research plan in 10 weeks? Like there's a lot of, how do you prioritize which experiments to do? It's like, there's a lot of inductive biases that go into that, that I don't like a language model would not do well at tha…”

“The autoregressive nature of language models means that if they make a mistake, and you correct it, and then say, no, that was a mistake, please do it this way instead. The more often you do that, the worse the dialogue answers get. Because it's in the trainin…”

“Language models are just that, right? They're models of language. They are not particularly well suited for understanding arbitrary, you know, like big numeric data sets.”

“I would imagine you're taking these different types of virtual cell models and eventually merging them into the equivalent of like a biological Omni model, kind of like how on the language model side, you had people that did language and then, you know, people…”

“I would describe it more as not unlearning, but maybe suppression. I think there's, like, really, like, I guess, guarantees that you've fully removed information from a model is, is, I don't think it's been convincingly showed anywhere yet”

“You can actually see, like the way that we, like, disentangle memorization, you can kind of see this like, gradient of memorization in between both mechanistically and behaviorally with, like, logical reasoning tasks being quite distinct from rote memorization…”

“Language models themselves are more like compute or GPU a generation ago, where what can we build at the layer above? And in software systems, we've traditionally thought of VMware being a great example. You have the operating system and the underlying archite…”

“With language models, when you get features, you still have to do auto interpret and things like that to actually get an understanding of what this concept is. But in image and video and world, it's like extremely easy to grok what the concept is because you c…”

“We've seen that models internally have some awareness of like uncertainty or some sort of like user pleasing behavior that leads to hallucinatory behavior.”

“They, in terms of parameters, are actually not very big. They are definitely below a billion parameters. You know, if you're here these days in LLM space, you know, a model with less than a billion parameters, you'd think can't do anything. But on the other ha…”

“Most of the people were actually not very good at prompting. Actually, language models have a better sense of how to prompt AI models. AI models know AI models better.”

“I think one thing pretty, pretty interesting. I think might be happening soon is the language models will be like context aware and manage its own context.”

“So surprisingly video models is like the cost is very, is comparable to language models. And obviously the largest scale is language model. Maybe like a medium scale language models.”

“Video agents, mostly language models, they'll call these generative model, either it's a separate model or a diffusion head or whatever as tool. So this model can iteratively Refine the results or even like you generate longer content through a very long trend…”

“The visual intelligence are actually mostly coming from language. Like, these video models, especially from now, since the diffusion model technology is more mature, the, like, every time you see there, there's some improvement on these models, I would say mos…”

“In language models, it is roughly 100 to a thousand times every 12 to 18 months for the same given level of LMSYS ELO.”

“Yes, you can do a lot of hypothesis generation. You can have ideas, but ideas are not enough, right? So you can have a lot of ideas. The bottleneck is going, testing, and verifying that they work in the real world.”

“A lot of regulatory frameworks equate AI with language models and Yes, language models can, you know, manipulate people, can have all these kinds of harmful impacts that we should think about controlling, but AI for science is different. So I think this one si…”

“And the difference there is compared to language where self-improvement needs something like human feedback or other reward signals that are very sparse. They just tell you yes or no, thumbs up or down. We have dense feedback because the physics laws, there's …”