Royzen: Training on code unlocked general spatial and temporal reasoning
Michael Royzen · Beating GPT-4 with Open Source Models - with Michael Royzen of Phind · Nov 3, 2023 · at 43:51
Michael Royzen, CEO of Phind, shares findings that fine-tuning language models on code improves non-coding logical reasoning and world-modeling capabilities.
0:00 / 0:32exact quote · 32.6s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“We've seen emerging capabilities in the find model, whereby training it on high quality code, it can actually, like, reason better. It went from not being able to solve like, World problems where like riddles where like with like temporal and like low, like placement of objects and moving and stuff like that that GPT-IV can do pretty well. We went from not being able to do those at all to being able to do them just by training on more code, which is wild.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Michael Royzen
Royzen: AI dev tools do not need to own the IDE
“Somewhere where I disagree with him is that you need to own the IDE. I think like he made kind of some good points about, you know, not having platform risk in the long term, but some of the, you know, features that were mentioned, like suggesting diffs, for e…”
Royzen: Fine-tuned open source models will beat proprietary in 2024
“So I think that even if a delta exists, in twenty-twenty-four, the delta between proprietary and open source won't be large enough that a startup like us, with a lot of data that we've collected, can take the data that we have, fine-tune an open source model, …”
Royzen: GPT-4 was trained on HumanEval, proving data contamination
“GPT-IV itself has been trained on human eval, and we know this because GPT-IV is able to predict the exact doc string in many of the problems. I've seen it predict, like, the specific example values in the doc string, which is extremely improbable for it to ju…”
Royzen: Phind built the first internet-scale LLM RAG search in 2022
“And to the best of my knowledge, I think that's the first example that I'm aware of a LLM search engine model that's effectively connected to, like, a large enough index that I would consider, like, an internet scale. So, so I think we were the first to releas…”
Phind CEO: Future Programming Will Just Be Problem Solving, Delegating Implementation to AI
“In the future, you know, in the future, I think programming is just going to be really just the problem solving. Like you come up with an idea, you come up with like the basic design for the algorithm in your head, and you just tell the AI, hey, just like, jus…”
Royzen: Users switch to Phind when ChatGPT-4 fails on code
“What really shocks us is that a lot of the people who do that they're coming from ChatGPT. So they tried it in ChatGPT with ChatGPT-IV. It didn't work. Maybe it required like some multi-step reasoning. Maybe it required to like, Some internet context or someth…”