High Signal Podcasts Evidence ledger
Method
Browse
How the evidence works
← All source episodes

Latent Space / episode intelligence

Code Interpreter == GPT 4.5 (w/ Simon Willison, Alex Volkov, Aravind Srinivas, Alex Graveley, et al.)

10 Jul 2023 33 published claims 6 attributable people
Open original episode ↗ Read the evidence

Speakers in the public record

Alex VolkovSimon WillisonShawn WangDaniel WilsonAl ChangAravind Srinivas

Claim mix

belief 17commitment 6evaluation 4recommendation 3preference 2uncertainty 1

Evidence policy

Every row below preserves an exact excerpt. Identified speakers are linked; unresolved voices are labeled and excluded from people counts.

Claim ledger

The useful parts, with receipts.

33 published records

05 / belief

Alright. I think, I think of this like very happy note. I, I saw some folks come up and folks, we've been at this for two and a half hours almost.

“Alright. I think, I think of this like very happy note. I, I saw some folks come up and folks, we've been at this for two and a half hours almost.”
Speaker
Not verified from transcript
Publisher
Latent Space

06 / belief

Like, so you just kind of go through the libraries and like think about what each library does which is kind of something I was doing last night using, I was trying to use chatty PC to annotate itself and just kind of enumerate its capabilities. So like, yeah you can run open cv, I think.

“Like, so you just kind of go through the libraries and like think about what each library does which is kind of something I was doing last night using, I was trying to use chatty PC to annotate itself and just kind of enumerate its capabilities. So like, yeah you can run open cv, I think.”
Speaker
Shawn Wang
Publisher
Latent Space

07 / belief

I think the, the process that you're describing that people are, who are, have been using it more actively or describing that's like, it's like the most valuable training data in the entire world is like you know, because before you know, when you would generate some code with either Codex or with or whatever open, I didn't know if the code worked, whereas now you're, they're, they're getting signals from that because they're trying to run it and they're critiquing it and then regenerating it, and then you, maybe you're critiquing it.

“I think the, the process that you're describing that people are, who are, have been using it more actively or describing that's like, it's like the most valuable training data in the entire world is like you know, because before you know, when you would generate some code with either Codex or with or whatever open, I didn't know if the code worked, whereas now you're, they're, they're getting signals from that because they're trying to run it and they're critiquing it and then regenerating it, and then you, maybe you're critiquing it.”
Speaker
Not verified from transcript
Publisher
Latent Space

08 / commitment

I find that for the kind of stuff I'm doing, normally what I'll do is I'll take the code I'm working on, I will reduce it to the shortest sort of example that, that shows what I'm trying to do to use less tokens, copy and paste that in.

“I find that for the kind of stuff I'm doing, normally what I'll do is I'll take the code I'm working on, I will reduce it to the shortest sort of example that, that shows what I'm trying to do to use less tokens, copy and paste that in.”
Speaker
Simon Willison
Publisher
Latent Space

09 / belief

I agree with you though because the interface, the CHE interface doesn't let us access any temperatures and so we randomly get like random stuff.

“I agree with you though because the interface, the CHE interface doesn't let us access any temperatures and so we randomly get like random stuff.”
Speaker
Not verified from transcript
Publisher
Latent Space

10 / belief

I think I need to be able to provide an API key and say, here's the endpoint and here's what my data looks like, and you can query my data directly and and analyze it.

“I think I need to be able to provide an API key and say, here's the endpoint and here's what my data looks like, and you can query my data directly and and analyze it.”
Speaker
Not verified from transcript
Publisher
Latent Space

11 / belief

One thing I would say is that if you can upload up to a hundred megabytes, a hundred megabyte file, if you, you can, for most business on six problems, you can get that down to less than a hundred megabytes of data.

“One thing I would say is that if you can upload up to a hundred megabytes, a hundred megabyte file, if you, you can, for most business on six problems, you can get that down to less than a hundred megabytes of data.”
Speaker
Simon Willison
Publisher
Latent Space

12 / belief

I think, I think given that we're here talking about code interpreters specifically, I wanna just like summarize the point about code interpreters.

“I think, I think given that we're here talking about code interpreters specifically, I wanna just like summarize the point about code interpreters.”
Speaker
Not verified from transcript
Publisher
Latent Space

13 / belief

Like I want, like I think I really wanna see something where you can just take some documentation, like some documentation from like a website and then pull that and then utilize the examples from that documentation or whatever.

“Like I want, like I think I really wanna see something where you can just take some documentation, like some documentation from like a website and then pull that and then utilize the examples from that documentation or whatever.”
Speaker
Daniel Wilson
Publisher
Latent Space

14 / evaluation

I mean, the idea is basically that it goes down this line of thinking and it can't kind of back up, right? So it has to sort of finish its thought and then once it's finished its thought, it's like, oh wait, this doesn't work.

“I mean, the idea is basically that it goes down this line of thinking and it can't kind of back up, right? So it has to sort of finish its thought and then once it's finished its thought, it's like, oh wait, this doesn't work.”
Speaker
Al Chang
Publisher
Latent Space

15 / commitment

I will just point out before you stepped in we have Simon here who some people can say Simon knows how to code, and Simon also highlighted some things that he did that he wouldn't otherwise be able to do.

“I will just point out before you stepped in we have Simon here who some people can say Simon knows how to code, and Simon also highlighted some things that he did that he wouldn't otherwise be able to do.”
Publisher
Latent Space

17 / belief

I think Simon mentioned in the beginning, I was I'm not sure if you were here, that it, it was possible to upload like egg files or, or wheel files and then extend the Python can runtime and now it's no, no longer possible.

“I think Simon mentioned in the beginning, I was I'm not sure if you were here, that it, it was possible to upload like egg files or, or wheel files and then extend the Python can runtime and now it's no, no longer possible.”
Speaker
Not verified from transcript
Publisher
Latent Space

22 / commitment

I, I, I I will say this one thing almost a year ago, Jesus, it's been a year when Stable Diffusion released just before it released, there was like a better stable diffusion discord.

“I, I, I I will say this one thing almost a year ago, Jesus, it's been a year when Stable Diffusion released just before it released, there was like a better stable diffusion discord.”
Speaker
Not verified from transcript
Publisher
Latent Space

24 / belief

I think there's a, I think there's a fine tuned model under it, which, if that's the case, then replicating it using functions is gonna be pretty difficult.

“I think there's a, I think there's a fine tuned model under it, which, if that's the case, then replicating it using functions is gonna be pretty difficult.”
Speaker
Simon Willison
Publisher
Latent Space

26 / recommendation

If you look up on top on the jumbotron, I think one of the first tweets there, there's a, a quick video for those of you, but like, if you don't want to just go to settings and ChatGPT, go to Beta features and enable code interpreter, just hit the, the little, little toggle there and you'll have access under GPT four, you'll have access to a new code interpreter able , which does amazing things.

“If you look up on top on the jumbotron, I think one of the first tweets there, there's a, a quick video for those of you, but like, if you don't want to just go to settings and ChatGPT, go to Beta features and enable code interpreter, just hit the, the little, little toggle there and you'll have access under GPT four, you'll have access to a new code interpreter able , which does amazing things.”
Speaker
Alex Volkov
Publisher
Latent Space

27 / recommendation

Let us like use it via the api, but have our own function that de evaluates code, because then I can build my dream version of code interpreter with.

“Let us like use it via the api, but have our own function that de evaluates code, because then I can build my dream version of code interpreter with.”
Speaker
Simon Willison
Publisher
Latent Space

28 / recommendation

Okay. It's best if we don't tell them to the open AI folks because they'll just add them as more instructions to the moderation engine and change the model soon.

“Okay. It's best if we don't tell them to the open AI folks because they'll just add them as more instructions to the moderation engine and change the model soon.”
Speaker
Daniel Wilson
Publisher
Latent Space

29 / commitment

Well, well because, so my day job, my, my, my principal project, I ru, I run this open source project called Dataset, which is all about building tools to help people interrogate their data.

“Well, well because, so my day job, my, my, my principal project, I ru, I run this open source project called Dataset, which is all about building tools to help people interrogate their data.”
Speaker
Simon Willison
Publisher
Latent Space

30 / preference

I don't want to spend five or 10 minutes tediously debugging my way through it, through a, a nested list algorithm, but I'm happy for code interpreter to, to go ahead and do that. And in fact, when it wrote the code, I watched it make the exact same mistakes I would've made, like forgetting, getting off one off by one errors and all that kind of thing.

“I don't want to spend five or 10 minutes tediously debugging my way through it, through a, a nested list algorithm, but I'm happy for code interpreter to, to go ahead and do that. And in fact, when it wrote the code, I watched it make the exact same mistakes I would've made, like forgetting, getting off one off by one errors and all that kind of thing.”
Speaker
Simon Willison
Publisher
Latent Space

32 / evaluation

Yes. So in that case, that means that some of the thick barriers I've been running to are more the model being told no, pretend that you can't do it.

“Yes. So in that case, that means that some of the thick barriers I've been running to are more the model being told no, pretend that you can't do it.”
Speaker
Simon Willison
Publisher
Latent Space

33 / evaluation

Because it, it just, it gives you so many capabilities that ChatGPT and even ChatGPT with plugins doesn't really touch you on, especially if you know how to use it.

“Because it, it just, it gives you so many capabilities that ChatGPT and even ChatGPT with plugins doesn't really touch you on, especially if you know how to use it.”
Speaker
Simon Willison
Publisher
Latent Space
Search evidence