compute per token

1 statements across 1 episodes · 0 bullish · 0 bearish · 1 people on the record · first statement Jul 23, 2024 by Thomas Scialom · across every show →

Everything said about compute per token, oldest first

Jul 23, 2024 neutral
Insight
Scialom: Fixed compute per token in transformers forces models to generate extra tokens to think
“We are lacking of flexibility for pre-training architecture, transformers, where We spend the same amount of compute per token. And so, because of that, how can you, like, mitigate this? By generating more tokens, so more thoughts, more compute, because you ha…”
Thomas Scialom Jul 23, 2024 ▶ 51:19 Training Llama 2, 3 & 4: The Path to Open Source AGI — with Thomas Scialom of Meta AI
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.