Efficient Streaming Language Models with Attention Sinks (idlemachines.co.uk)

2 points by smaddrellmander 1d ago ↗ HN

0 comments

[ 3.3 ms ] story [ 9.9 ms ] thread

No comments yet.

A showcase of shard-db. Data from the Hacker News API. Refreshed every 15 minutes.

Source · Live DB stats