CARBS

3 statements across 2 episodes · 2 bullish · 0 bearish · 2 people on the record · first statement Oct 21, 2023 by Kanjun Qiu · said 18 times in 2 episodes since 2023 · across every show →

Mentions by year

brought up most by Kanjun Qiu (4), Josh Albrecht (3)

tap a year for its mentions
008115120232024episodesmentions
01120232024episodes it came up in
007.50.515120232024episodesmentions per episode
2024 14 mentions in 1 episode
2023 4 mentions in 1 episode

every mention, scene by scene, with the transcript →

Everything said about CARBS, oldest first

Oct 21, 2023 neutral
Assertion Not publicly verifiable
Imbue's CARBS hyperparameter optimizer originated from plasma physics concepts
“CARBS, our hyperparameter optimizer, came from Abe trying to automate his own research process doing hyperparameter optimization, and he actually pulled some ideas from plasma physics, he's a plasma physicist, to make the local search work.”
Kanjun Qiu Oct 21, 2023 ▶ 16:39 Why AI Agents Don't Work (yet) - with Kanjun Qiu of Imbue
Jun 25, 2024 positive
Insight
Albrecht: CARBS models compute cost per sample for hyperparameter search
“CARBS is, it's maybe a backronym, but it's for Cost Aware Pareto Region Bayesian Search... The point is that it's a cost aware hyperparameter tuner. So most hyperparameter tuners you kind of say, okay, here's this objective function. I want you to make this nu…”
Josh Albrecht Jun 25, 2024 ▶ 47:25 State of the Art: Training 70B LLMs on 10,000 H100 clusters
Jun 25, 2024 positive
Insight
Albrecht: Cost-aware tuning reveals scaling laws for all hyperparameters
“So by doing that, we can see the scaling laws or not just, you know, the scaling laws from like the, you know, chinchilla paper, the scaling laws for all parameters. We can see how does the number of layers change with this? How does the You know, the learning…”
Josh Albrecht Jun 25, 2024 ▶ 48:50 State of the Art: Training 70B LLMs on 10,000 H100 clusters
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 200 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.