Ring Attention

topic on 1 show · 2 statements across 1 episodes

2 statements about Ring Attention, every show

LATENT SPACE Assertion Not checkable as stated

Huang: EasyContext was the first functional PyTorch Ring Attention implementation

“Easy context was the first PyTorch implementation that applied it with native libraries that worked pretty well. And then we adapted it ourselves in order to configure it for our cluster network topology.”

Mark Huang May 31, 2024 ▶ 27:08 How to train a Million Context LLM — with Mark Huang of Gradient.ai

LATENT SPACE Assertion Not checkable as stated

Huang: Berkeley's JAX Ring Attention does not work well on GPUs

“The Jaxx implementation just does not work on, on GPUs very well. Like, any naive setup that you do, like, it just won't run out of the box very easily”

Mark Huang May 31, 2024 ▶ 26:02 How to train a Million Context LLM — with Mark Huang of Gradient.ai

← every entity, every show

Turn any episode into a week of clips.

This entire site, thousands of episodes across every show 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.

Ring Attention

2 statements about Ring Attention, every show

Turn any episode into a week of clips.