char-fork/llama.cpp

git clone https://git.t4t.associates/char-fork/llama.cpp

Charlotte Somenc/dec in server context8300f1411

char/aliceai-t5-moe
2 folders159 files

files

template-instances/
vendors/
CMakeLists.txt10.5 KiB
acc.cu2.4 KiB
acc.cuh131 B
add-id.cu1.9 KiB
add-id.cuh101 B
allreduce.cu44.2 KiB
allreduce.cuh1014 B
arange.cu1.2 KiB
arange.cuh137 B
argmax.cu2.8 KiB
argmax.cuh98 B
argsort.cu12.8 KiB
argsort.cuh950 B
binbcast.cu24.2 KiB
binbcast.cuh672 B
clamp.cu1.6 KiB
clamp.cuh135 B
col2im-1d.cu3.0 KiB
col2im-1d.cuh104 B
common.cuh63.8 KiB
concat.cu9.5 KiB
concat.cuh137 B
conv-transpose-1d.cu3.3 KiB
conv-transpose-1d.cuh158 B
conv2d-dw.cu7.2 KiB
conv2d-dw.cuh155 B
conv2d-transpose.cu4.7 KiB
conv2d-transpose.cuh158 B
conv2d.cu6.4 KiB
conv2d.cuh149 B
convert.cu25.4 KiB
convert.cuh2.3 KiB
count-equal.cu2.0 KiB
count-equal.cuh144 B
cp-async.cuh2.1 KiB
cpy-utils.cuh6.2 KiB
cpy.cu30.8 KiB
cpy.cuh227 B
cross-entropy-loss.cu5.9 KiB
cross-entropy-loss.cuh251 B
cumsum.cu10.7 KiB
cumsum.cuh137 B
dequantize.cuh17.2 KiB
diag.cu2.6 KiB
diag.cuh133 B
diagmask.cu1.7 KiB
diagmask.cuh150 B
dsv4-hc.cu9.5 KiB
dsv4-hc.cuh292 B
fattn-common.cuh47.4 KiB
fattn-mma-f16.cuh105.4 KiB
fattn-swizzle.cuh5.0 KiB
fattn-tile.cu2.1 KiB
fattn-tile.cuh61.7 KiB
fattn-vec.cuh24.4 KiB
fattn.cu28.2 KiB
fattn.cuh271 B
fill.cu1.1 KiB
fill.cuh99 B
fwht.cu3.1 KiB
fwht.cuh192 B
gated_delta_net.cu13.7 KiB
gated_delta_net.cuh689 B
getrows.cu22.6 KiB
getrows.cuh622 B
ggml-cuda.cu231.1 KiB
gla.cu3.0 KiB
gla.cuh112 B
im2col.cu12.4 KiB
im2col.cuh218 B
lightning-indexer.cu23.6 KiB
lightning-indexer.cuh190 B
mean.cu3.0 KiB
mean.cuh99 B
mma.cuh62.0 KiB
mmf.cu7.7 KiB
mmf.cuh44.6 KiB
mmid.cu8.9 KiB
mmid.cuh278 B
mmq-config-ampere.cuh35.4 KiB
mmq-config-blackwell.cuh3.5 KiB
mmq-config-cdna.cuh15.8 KiB
mmq-config-gcn.cuh25.6 KiB
mmq-config-pascal-dp4a.cuh24.5 KiB
mmq-config-pascal-older.cuh24.5 KiB
mmq-config-rdna2.cuh24.8 KiB
mmq-config-rdna3-5.cuh26.6 KiB
mmq-config-rdna3.cuh24.8 KiB
mmq-config-rdna4.cuh26.6 KiB
mmq-load-tiles.cuh79.2 KiB
mmq-vec-dot.cuh47.1 KiB
mmq.cu15.8 KiB
mmq.cuh70.2 KiB
mmvf.cu39.8 KiB
mmvf.cuh826 B
mmvq.cu71.2 KiB
mmvq.cuh988 B
moe-weighted-reduction.cu3.4 KiB
moe-weighted-reduction.cuh411 B
norm.cu29.9 KiB
norm.cuh838 B
opt-step-adamw.cu2.9 KiB
opt-step-adamw.cuh150 B
opt-step-sgd.cu1.7 KiB
opt-step-sgd.cuh146 B
out-prod.cu4.8 KiB
out-prod.cuh100 B
pad.cu4.4 KiB
pad.cuh131 B
pad_reflect_1d.cu3.4 KiB
pad_reflect_1d.cuh153 B
pool1d.cu2.6 KiB
pool1d.cuh137 B
pool2d.cu3.2 KiB
pool2d.cuh137 B
quantize.cu29.2 KiB
quantize.cuh3.4 KiB
reduce_rows.cuh1.2 KiB
roll.cu2.4 KiB
roll.cuh133 B
rope.cu43.5 KiB
rope.cuh486 B
scale.cu1.5 KiB
scale.cuh135 B
set-rows.cu15.0 KiB
set-rows.cuh155 B
set.cu1.2 KiB
set.cuh145 B
snake.cu2.8 KiB
snake.cuh366 B
softcap.cu1.4 KiB
softcap.cuh158 B
softmax.cu17.3 KiB
softmax.cuh228 B
solve_tri.cu11.2 KiB
solve_tri.cuh104 B
ssm-conv.cu9.0 KiB
ssm-conv.cuh176 B
ssm-scan.cu40.6 KiB
ssm-scan.cuh103 B
sum.cu1.3 KiB
sum.cuh210 B
sumrows.cu2.3 KiB
sumrows.cuh212 B
top-k.cu10.2 KiB
top-k.cuh100 B
topk-moe.cu16.9 KiB
topk-moe.cuh1.1 KiB
tri.cu5.2 KiB
tri.cuh131 B
tsembd.cu1.8 KiB
tsembd.cuh161 B
unary.cu24.7 KiB
unary.cuh4.5 KiB
upscale.cu13.1 KiB
upscale.cuh139 B
vecdotq.cuh48.6 KiB
wkv.cu9.4 KiB
wkv.cuh218 B