AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
5min chapter
Parallelism and Acceleration for Large Language Models with Bryan Catanzaro - #507
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Scaling Large Language Models to Large Cepe Clusters
Aly: We have a paper at the super computing conference coming up in november on scaling very large language models to large cepe clusters with many thousand ceps. And that's of our magatron project, we want to be able to train superassum super big transformers on h pc infor structure. When we're training very large gpt three style models on our d g x super pod, we are actually sustaining 52 % of the tenser peke through put over the entire train run. Aly sounds like there is plan to bein imperson component to the conference. Er wouldn't that be fantastic? I mean, but in terms of megatron, tell us a
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode