fix(models): preserve lm-head dtype boundaries (#3491)
* fix(models): preserve lm-head dtype boundaries Signed-off-by: larkzhang-nv <larkz@nvidia.com> * fix(models): infer lm-head compute dtype Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com> * test(models): cover NeMo RL lm-head dtypes Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com> --------- Signed-off-by: larkzhang-nv <larkz@nvidia.com> Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com> Co-authored-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
J
jQizhang committed
42af068ccab01c2e7166f96a4e1886e63e66e863
Parent: 1cd4fd0
Committed by GitHub <noreply@github.com>
on 8/12/2026, 2:23:15 AM