Chris Olah
speaker
254 appearances
1 recordings
1 series
first heard Nov 2024
last heard Nov 2024
Chris Olah’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
And so it's like this desire for extreme clarity. So it's like anyone could just pick up your paper, read it and know exactly what you're talking about. It's why it can almost be kind of dry. Like all of the terms are defined. Every objection's kind of gone through methodically.
And it makes sense to me because I'm like, when you're in such an a priori domain, clarity is sort of this way that you can prevent people from just kind of making stuff up. And I think that's sort of what you have to do with language models. Like very often I actually find myself doing sort of mini versions of philosophy. You know, so I'm like, suppose that you give me a task.
I have a task for the model and I want it to like pick out a certain kind of question or identify whether an answer has a certain property. Like I'll actually sit and be like, let's just give this a name, this property. So like, you know, suppose I'm trying to tell it like, oh, I want you to identify whether this response was rude or polite.
I'm like, that's a whole philosophical question in and of itself. So I have to do as much philosophy as I can in the moment to be like, here's what I mean by rudeness and here's what I mean by politeness. And then there's another element that's a bit more, I guess... I don't know if this is scientific or empirical. I think it's empirical.
So I take that description and then what I want to do is, again, probe the model many times. Prompting is very iterative. I think a lot of people, if a prompt is important, they'll iterate on it hundreds or thousands of times. And so you give it the instructions and then I'm like, what are the edge cases?
So if I looked at this, so I try and like almost like, you know, see myself from the position of the model and be like, what is the exact case that I would misunderstand or where I would just be like, I don't know what to do in this case. And then I give that case to the model and I see how it responds. And if I think I got it wrong, I add more instructions or even add that in as an example.
So these very like taking the examples that are right at the edge of what you want and don't want. and putting those into your prompt as like an additional kind of way of describing the thing. And so yeah, in many ways, it just feels like this mix of like, it's really just trying to do clear exposition. And I think I do that because that's how I get clear on things myself.
So in many ways, like clear prompting for me is often just me understanding what I want. It's like half the task.
Yeah, I think that prompting does feel a lot like the kind of the programming using natural language and experimentation or something. It's an odd blend of the two. I do think that for most tasks, so if I just want Claude to do a thing, I think that I am probably more used to knowing how to ask it to avoid like common pitfalls or issues that it has. I think these are decreasing a lot over time.
But it's also very fine to just ask it for the thing that you want. I think that prompting actually only really becomes relevant when you're really trying to eke out the top like 2% of model performance. So for like a lot of tasks, I might just, you know, if it gives me an initial list back and there's something I don't like about it, like it's kind of generic.
Like for that kind of task, I'd probably just take a bunch of questions that I've had in the past that I've thought worked really well and I would just give it to the model and then be like, now here's this person that I'm talking with. give me questions of at least that quality. Or I might just ask it for some questions.
And then if I was like, oh, these are kind of trite or like, you know, I would just give it that feedback and then hopefully it produces a better list. I think that kind of iterative prompting At that point, your prompt is like a tool that you're going to get so much value out of that you're willing to put in the work.
Like if I was a company making prompts for models, I'm just like, if you're willing to spend a lot of like time and resources on the engineering behind like what you're building, then the prompt is not something that you should be spending like an hour on. It's like that's a big part of your system. Make sure it's working really well. And so it's only things like that.
Like if I, if I'm using a prompt to like classify things or to create data, that's when you're like, it's actually worth just spending like a lot of time, like really thinking it through.
You know, there's a concern that people over-anthropomorphize models, and I think that's a very valid concern. I also think that people often under-anthropomorphize them, because sometimes when I see issues that people have run into with Claude, you know, say Claude is refusing a task that it shouldn't refuse, but then I look at the text and the specific wording of what they wrote, and I'm like...
I see why Claude did that. And I'm like, if you think through how that looks to Claude, you probably could have just written it in a way that wouldn't evoke such a response. Especially this is more relevant if you see failures or if you see issues. It's sort of like, think about what the model failed at. Like, what did it do wrong?
and then maybe it gave that will give you a sense of like why um so is it the way that i phrased the thing and obviously like as models get smarter you're going to need less in this less of this and i already see like people needing less of it but that's probably the advice is sort of like try to have sort of empathy for the model like read what you wrote as if you were like a kind of like person just encountering this for the first time how does it look to you and what would have made you behave in the way that the model behaved so if it misunderstood what kind of like
what coding language you wanted to use. Is that because like, it was just very ambiguous and it kind of had to take a guess in which case next time you could just be like, Hey, make sure this is in Python or, I mean, that's the kind of mistake I think models are much less likely to make now. But you know, if you, if you do see that kind of mistake, that's, that's probably the advice I'd have.
Yeah, I mean, I've done this with the models. It doesn't always work, but sometimes I'll just be like, why did you do that? I mean, people underestimate the degree to which you can really interact with models. And sometimes I'll just quote word for word the part that made you... And you don't know that it's fully accurate, but sometimes you do that and then you change a thing.
I mean, I also use the models to help me with all of this stuff, I should say. Prompting can end up being a little factory where... You're actually building prompts to generate prompts. And so like, yeah, anything where you're like having an issue, asking for suggestions, sometimes just do that. Like you made that error. What could I have said? That's actually not uncommon for me to do.
Showing 61–80 of 254 · page 4 of 13
← Previous
Next →