cuda : add RoPE kernel for mode == 2 (NeoX) #2760
cuda : add RoPE kernel for mode == 2 (NeoX)
ac4bb6ba
falcon : do not offload the embeddings layer
333e27b3
ggerganov
merged
3f460a2b
into master 2 years ago
ggerganov
deleted the fix-falcon-cuda branch 2 years ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub