1
0
slaren 0d56246f4b ggml : group all experts in a single ggml_mul_mat_id (#6505) 1 жил өмнө
..
acc.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
acc.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
alibi.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
alibi.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
arange.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
arange.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
argsort.cu 08a0c02060 ggml : mul_mat_id use the same tensor for all the experts (#6387) 1 жил өмнө
argsort.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
binbcast.cu 0d56246f4b ggml : group all experts in a single ggml_mul_mat_id (#6505) 1 жил өмнө
binbcast.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
clamp.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
clamp.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
common.cuh 5dc9dd7152 llama : add Command R Plus support (#6491) 1 жил өмнө
concat.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
concat.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
convert.cu 0d56246f4b ggml : group all experts in a single ggml_mul_mat_id (#6505) 1 жил өмнө
convert.cuh 5dc9dd7152 llama : add Command R Plus support (#6491) 1 жил өмнө
cpy.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
cpy.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
dequantize.cuh 5dc9dd7152 llama : add Command R Plus support (#6491) 1 жил өмнө
diagmask.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
diagmask.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
dmmv.cu 5dc9dd7152 llama : add Command R Plus support (#6491) 1 жил өмнө
dmmv.cuh d48ccf3ad4 sync : ggml (#6351) 1 жил өмнө
getrows.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
getrows.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
im2col.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
im2col.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
mmq.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
mmq.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
mmvq.cu 55c1b2a3bb IQ1_M: 1.75 bpw quantization (#6302) 1 жил өмнө
mmvq.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
norm.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
norm.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
pad.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
pad.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
pool2d.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
pool2d.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
quantize.cu 5dc9dd7152 llama : add Command R Plus support (#6491) 1 жил өмнө
quantize.cuh 5dc9dd7152 llama : add Command R Plus support (#6491) 1 жил өмнө
rope.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
rope.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
scale.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
scale.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
softmax.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
softmax.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
sumrows.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
sumrows.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
tsembd.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
tsembd.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
unary.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
unary.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
upscale.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
upscale.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 жил өмнө
vecdotq.cuh 55c1b2a3bb IQ1_M: 1.75 bpw quantization (#6302) 1 жил өмнө