Latent Space / episode intelligence
Speakers in the public record
Claim mix
Evidence policy
Every row below preserves an exact excerpt. Identified speakers are linked; unresolved voices are labeled and excluded from people counts.
Claim ledger
10 published records
01 / uncertainty
“I believe, Hermes was, is like a quarter million rows of data, I don't know the actual byte size on that particular one.”
02 / uncertainty
“Being part of AI Twitter is seeing all these things and going dude, I'm being DDoS'ed by all these things and I don't know how different they are.”
03 / evaluation
“yeah. So, we'll see, but I wanted to highlight the speed because you started with like within a week the first alpaca or, implementation and change in Axelot came and now like you're talking about like three days and that's with you flying and that's with you like presenting and talking on podcasts.”
04 / evaluation
“It tended to be, like, have relatively good answers, depending on who you ask, right? But, I think for me, it was just, Releasing models was a way to try and, like, continue to build out the product, figure out what I needed to put into the product, how do I make it faster, and, if you've got to, like, go and debug your product, you may as well have it do something useful.”
05 / preference
“He's amazing, I love him, and I think everybody should like, show him some love, and then, but yeah, like, ultimately, the, I think the, the big, yeah, the biggest thing that I could ask for would be just, yeah, more core contributors.”
06 / preference
“Like, so the first thing on the roadmap is to write the roadmap, and then sort of going from there, it's, I think, So for me the sort of the vision is like it's it's a developer first platform right and as a developer You you're maybe you're more than likely doing it this sort of this side hustle side project trying to figure out like how do I build?”
07 / evaluation
“Just because, you've quantized these down, so your accuracy is probably a little off, and so that by the time you've done the Qlora, you're not moving the weights how you would on a full fine tune with the full parameter weights.”
08 / evaluation
“Nobody liked it. It's really, so first of all, nobody liked it, but secondly, only the people who were able to find compute enough to run inference on this, they only had like, I can't run this on my laptop, and so that's why it's much easier, something like OpenRMS 7 to be, 7B, it's much easier, because you can run this on your MacBook.”
09 / evaluation
“3, 500 words that you're not using, right? And typically that's either filled with these PAD tokens, so I think I made the analogy last night that it's like having sort of like a glass here you fill it up with a shot of liquor and then you're and that's your training data and then you just fill it up with more water and those are your PAD tokens and it's just, it doesn't do much, right?”
10 / recommendation
“Yeah, it is, and I guess like, if you really wanted to, like, figure it out, I guess you could, like, git blame everything, and But, yeah, it's, so, I think that's always a useful thing, it's like Because people want to experiment but they don't, people will get frustrated when you've experiment, you're experimenting and it breaks and you don't know why or you know why and you've just gone down the rabbit hole, right? So, so I think that's one of the big features that's, that I think I find important because it's It prevents you from doing things you probably shouldn't have, and it, and sometimes we will let you do those things, but we'll try and warn, warn you that you've done that.”