Evidence receipt / recommendation
Published · transcript-backedNathan Lambert: recommendation
19 Jul 2023 Latent Space Llama 2: The New Open LLM SOTA (ft. Nathan Lambert, Matt Bornstein, Anton Troynikov, Russell Kaplan, Whole Mars Catalog et al.)
“We like did a, like a nice experimentation of that hugging face and it's, it's out there, it's ready for someone to invest more time in it and do it.”
Source trail
Everything needed to verify it.
- Speaker
- Nathan Lambert
- Attribution
- Verified speaker
- Claim type
- recommendation
- Recorded
- 19 Jul 2023
- Publisher
- Latent Space
Transcript context
…stoked. Yeah. This, this conversation gave me a bit of an idea for essentially llama stack overflow. Like you, you imagine like Stack overflow with with like sort of llama at, its at its base, but then like, it's not very good at coding, but we can actually do ratings on like, you know, preference ratings on, on answers and, and, and entire conversation chains. And at, at some point, we'll, we'll accumulate the, the code DA dataset that we need to find here in lama. That would probably do it. Yeah, we, we've like, there's challenges in base models and how to execute code to get feedback and stuff, but, We've seen early experiments and like we've worked on one, funnily enough that was called Stack Lama. We like did a, like a nice experimentation of that hugging face and it's, it's out there, it's ready for someone to invest more time in it and do it. I think especially now that Llama2, I'm like, Lama two's gonna be easier to work with. It's just better language models are a little bit easier to steer. Absolutely. Alex, you have and Mars catalog you, you just joined and I I am sure you have a question. Yeah, go ahead Alex.…
Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.