llama.cpp
ggml-cuda : use int64 for norm forward dst offset (match rms_norm_back)
#24214
Closed
Closed
ggml-cuda : use int64 for norm forward dst offset (match rms_norm_back) #24214
palios-taey
wants to merge
1 commit into
ggml-org:master
from
palios-taey:norm-int64-offset
ggml-cuda : use int64 for norm forward dst offset (match rms_norm_back)
3dfce13e
Reviewers
No reviews
Assignees
No one assigned
Labels
Nvidia GPU
ggml
Milestone
No milestone
Login to write a write a comment.
Login via GitHub