Comment by dnhkng
1 month ago
Maybe, but the interesting thing for me it this only works with specific 'chunks' of the transformer layer stack. More or less that the optimal leads to worse performance.
1 month ago
Maybe, but the interesting thing for me it this only works with specific 'chunks' of the transformer layer stack. More or less that the optimal leads to worse performance.
No comments yet
Contribute on Hacker News ↗