Training vs. Memorization: The Jane Austen Thought Experiment
Is AI just a giant database of quotes? Not exactly.
Training vs. Memorization: The Jane Austen Thought Experiment
Is AI just a giant database of quotes? Not exactly.
In Lesson 3 of our Prompt Engineering module, we dive into a thought experiment: What happens if we train a language model only on the works of Jane Austen?
This video breaks down the crucial difference between memorization and statistical mapping. We explore why an AI trained on Austen doesn't just regurgitate her books, but instead learns the "probabilistic essence"—the rhythms, vocabulary, and social patterns—of her 19th-century world.
In this 10-minute lesson, we cover:
Statistical Parrots?: Why "predicting the next word" is a simple mechanic that leads to stunningly complex results.
The Training Data Bias: Why your model can't tell you about a "laptop" or "hip-hop" if it isn't in the training corpus.
Midjourney & Wonder Woman: A real-world look at how training data creates visual bias (e.g., Gal Gadot vs. Lynda Carter).
The Mini-Turing Test: How experts evaluate if a model has truly captured the "soul" of a writer's style.
Closed World vs. The Internet: Setting the stage for our next move—scaling these small patterns to the entire internet to create Large Language Models (LLMs).
Prompt Engineering Insight: Understanding that AI is a "prediction machine" helps us realize why it makes mistakes. As prompt engineers, our job is to navigate these statistical patterns to get the most accurate and useful results possible.
Transcript
A lightly formatted transcript of the video, with timestamps so you can jump to any section on YouTube.
[0:01] Okay, Mc Bear here again in our little experiment to try to figure out what language models are, what what that means exactly. We're going to try a little thought experiment. We're going to imagine that we trained a language model only on Jane Austin's works. And you could do this if you want. We may set up a lab doing this and see the results of it. But for now, we're just going to treat this as a thought experiment. what would this model output? So our parameters are this is a closed world. This is only Jane Austin's world. And so what it'll do is it won't memorize and regurgitate
[0:43] Jane Austin text. What it'll do is it'll learn her patterns of writing statistically. It will learn the words she uses, what words are related to other words. And if you ask it a question like write a story about something, it will sort of predict what she may have written based on the statistics of you know her writing her corpus her works. This is not memorization. This is a statistical mapping. It looks at covariance and looks at patterns within her data in her data. It makes predictions. If you ask it something out of its scope, like what is a laptop? It doesn't know. It's not its world.
[1:28] It's not its training thing. It's not because it's stupid. It's because that's not its training data. And all training data has bias. If I go to Medjourney and I type in, you know, a photo of Wonder Woman, it shows me Gail Gadub. Uh there is a Wonder Woman before Gail Gadub called Linda Carter. Doesn't show me Linda Carter. It shows me Gail GDOTe. What that means is Met Journey is trained on Gail Gdote, you know, uh, films because that's what it shows. It maps sort of Wonder Woman to Gail Gdote or if you do a, you know, a drawing of Wonder Woman, it will probably do a TC comic version of that.
[2:10] So, what these life models do is they predict tires. They're just, some people call them statistical parrots. that may be demeaning a little bit about how amazing they actually are. It's remarkable what they do given how they work. They are making predictions. They make very good predictions. It's it's sort of stunning how good they are at what they do given how they work because they're just making predictions. But they get better and better at making the predictions. But because they make predictions, they can make mistakes. They can predict things wrong. And we're going to get into that quite a bit. We're going to get into how do we know
[2:49] It did a good job. Sometimes is qualitative. Like here I made some images and we'll do this down the road. We'll do image prompting for midjourney image fringing for V3 for books for scientific research etc. This is what I imagine Jane Austin to look like. It probably looks nothing like Jane Austin maybe. uh has her I read she had I forgot what the thing called the thing on her head that is something I read uh I forgot what it's called a language model would know what it's called she would know what it's called too because it's in her world in her world if he trained it on her world her world would
[3:31] Not know what all laptop is so what these things are doing and they're amazing is there are pomolistic machines if trained on Jane Austin data It gets sort of the probabilistic essence, the covariance of what Jane Austin might have written. There's not identity. We're not looking up and following her fact. We're sort of generating a novel thing that she may have written. And they're remarkably good at this. They're remarkably good at I'm not a super expert in Jane Austin. The way you would evaluate that can it write like Jane Austin is you get Jane Austin experts. You would generate text and ask them, well, does the same feel like Jane
[4:16] Austin? If they're real experts, they would know everything she wrote. So, they know that she did not write this. But if you sort of knew Jane Austin and knew her feel a test would be, could you tell sort of generated Jane Austin from real and that's a turn test, which we'll talk about uh a bit, many turn tests. We'll get into that in this video. What the essence of this video is is these are statistical machines. They learn patterns. They don't do lookup. They don't do linearization. And the our next video we're going to talk about okay this is just the Jane Austin world. What if we take them to
[4:54] The internet world a scale the entire internet and that's what these quote unquote large language models are. So when we use the word large we're talking about big big data like internet scale data data which it cost you know 10 million 20 million to train just from the computing power to train the models but you could you could train your own Jane Austin models I don't know how much she's written but in computer world she hasn't written that much compared to what a computer can handle even if she written a thousand butts which I don't think so I think it's maybe like 10 text of whatever I don't know. I'm not an
[5:31] Expert in that. Um, you could definitely train a model on Jane Austin's corpus. We may do that as a experiment in this class. Technically a project you could do if you're a student in this class and it would be limited to the Jane Austin world. I certain that she'd never use the word laptop or hiphop or, you know, whatever or, you know, stuff even out of her world in England. Um, and so that's not in her world, but her patterns of writing, it's going to learn, it's going to learn her style of writing from the statistical coariance of the patterns and the fake language models, the large language models. We'll
[6:15] Do this at an internet wide scale. It'd be interesting. You could ask a large language model to restrict yourself just to Jane Austin's world. it will try its best that it'll probably creep in some stuff that wasn't in her world. You couldn't really get a large angler model to truly be her world unless you only trained it on data from her time. You right now you can go to Claw Jedi Chad PT and say restrict yourself only to Jane Austin's universe and answer these questions. It'll do its best. It'll try, but things will creep in that it doesn't fully understand because it's just looking at statistical patterns and it
[6:58] May throw in the word laptop. It might have Jane Austin use the word laptop. If you trained it purely only on Jane Austin data from scratch, it just doesn't know that exists. So, it's not going to predict that because it's not in its training data. So, in our next one, we're going to get into the next step scale large language models. with the same idea feature inundated off identifiers and it'll have different kinds of biases. Uh I have another class where we talk about how to detect bias and what to do about bias. My bias here is this is my bias of what I think that's Emma her
[7:39] Most famous character and my bias of what I think Jane Austin looks like. It could be and probably is way way off. But um it's also med journey bias because I use medjourney to generate these images. It clearly is trained on the left fashion images. I just had a critique of somebody who's signing you know Jane Austin looks like a supermodel. You know she didn't look like a supermodel. I don't know anybody really knows exactly how she looked because there's only sketches of drawings. Um but it's true. the the midjourney bias is to make everybody look like they're in luck. Use a different model, train on different
[8:18] Data, it will have a different look in a different field. We'll get into that because choosing the right language model for the advanced this could be part of the prompt engineering experience, but we won't get into that now. I'm already over time. I think I'm over eight minutes and I want to keep these to five. Certainly less than 10. So the major point of this is we're not memorizing. It's not if we train it on Jane Austin data, it would not just regurgitate and spit out like a database exact Jane Austin quotes. It would learn statistical patterns within Jane Austin's work and make a prediction
[8:59] Of what the next word would be given her previous history of writing. And that same model is going to go to basically large language models. It's going to do the same thing but a much greater scale. So it's getting the patterns and rhythms. It's learning it statistically the patterns and rhythms of of Jane Austin just like it learns the patterns and rhythms of the internet. But those patterns are much more complicated because they're the internet. They're not just one writer. Uh, and so if you wanted to truly do a Jane Austin generator, you should probably limit it to Jane Austin data, which you can do. It's not that much.
[9:38] But the idea, again, the idea of this video is not memorization. It's not just looking up and spitting out something. It's looking at patterns, then generating, predicting novel sequences based on the patterns that it's learned. Okay, that's it. We're almost at 10 minutes. I think at 10 minutes right now. So I'm just going to cut this. The next one we're going to talk now about the next scale large language models. It should do the same thing but now at scale. Okay. Take care.
