llama.cpp
cuda : add RoPE kernel for mode == 2 (NeoX)
#2760
Merged

cuda : add RoPE kernel for mode == 2 (NeoX) #2760

ggerganov merged 2 commits into master from fix-falcon-cuda
ggerganov
ggerganov cuda : add RoPE kernel for mode == 2 (NeoX)
ac4bb6ba
ggerganov ggerganov requested a review from slaren slaren 2 years ago
ggerganov ggerganov requested a review from JohannesGaessler JohannesGaessler 2 years ago
ggerganov
ggerganov commented on 2023-08-24
ggerganov
slaren
JohannesGaessler
JohannesGaessler commented on 2023-08-24
slaren
JohannesGaessler
ggerganov
slaren
ggerganov
slaren
ggerganov falcon : do not offload the embeddings layer
333e27b3
ggerganov
ggerganov ggerganov merged 3f460a2b into master 2 years ago
ggerganov ggerganov deleted the fix-falcon-cuda branch 2 years ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone