Remove post flattening CUDA clone()s, for 2% speedup in a 1x16 7B llama2#43
Closed
jaemzfleming wants to merge 1 commit intoVahe1994:mainfrom jaemzfleming:jf/remove-clones
+2-2
clone()s, for 2% speedup in a 1x16 7B llama2#43