15 B models

1 statements across 1 episodes · 1 bullish · 0 bearish · 1 people on the record · first statement Oct 29, 2024 by Ethan He · across every show →

Everything said about 15 B models, oldest first

Oct 29, 2024 positive

Assertion Supported

He: Upcycling a 15B model on 1T tokens yielded 4% MMLU gain

“On other scaling experiments, we tried on 15 B models upcycling and applied on one trillion tokens and achieved roughly about five percent improvement in terms of the validation loss and four percent improvement on MMLU.”

Ethan He Oct 29, 2024 ▶ 21:12 [Paper Club] Upcycling Large Language Models into Mixture of Experts

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.

15 B models

Everything said about 15 B models, oldest first

Turn any episode into a week of clips.