Reduce the memory usage of logits from O(context_length) to O(1)
Differential Revision: D61246566 Pull Request resolved: https://github.com/pytorch/executorch/pull/4688
M
Mengtao Yuan committed
11e8ed33bed08b4cb893ab8dda201f683017dee7
Parent: 6c26a87
Committed by GitHub <noreply@github.com>
on 8/23/2024, 2:03:04 AM