Radio and PodcastRadio and PodcastLive Radio & Podcasts
#131: FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness artwork
Technology

#131: FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

Misreading Chat

Apr 23, 202430:40Technology

CUDA で書かれた PyTorch 用カーネルに森田が玉砕しました。

About This Episode

#131: FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness is an episode from Misreading Chat. CUDA で書かれた PyTorch 用カーネルに森田が玉砕しました。 The episode was published on Apr 23, 2024 and runs for 30:40. Use the player to listen...

Podcast

This episode belongs to Misreading Chat.

Listen Online

Use the player on this page to stream the episode online.

Episode Details

Published Apr 23, 2024, 30:40 long, audio available.