Re:Astra can read and write weights to improve Qwen
by vikmals on Monday November 02, 2026 @11:53PM
(#50348234)
This article really caught my attention, until the end of the article when the author says that the math improvement was only archived because it fixed the model from looping. This should be in the tl;dr at the top!
[ Parent
| on HN ]
There are no comments on this comment yet.