inference efficiency
1 statements across 1 episodes · 1 bullish · 0 bearish · 1 people on the record · first statement Jul 23, 2024 by Thomas Scialom · across every show →
Everything said about inference efficiency, oldest first
Jul 23, 2024 bullish
Scialom: Overtrain models beyond Chinchilla optimal to minimize inference costs
“And so, to be compute efficient at inference time, it's much better to train it much longer training time, even if it's an effort, an additional effort, than to have a bigger model. That's what I call, like, I refer to the chinchilla trap, Not that Chinchilla …”