Tags
Tags give the ability to mark specific points in history as being important
master-6432aab
6432aabb
·
cuda : fix missing const qualifier in casts (#2027)
·
Jun 28, 2023
master-b922bc3
b922bc35
·
llama : remove shards weight file support (#2000)
·
Jun 28, 2023
master-7f9753f
7f9753fa
·
CUDA GPU acceleration for LoRAs + f16 models (#1970)
·
Jun 28, 2023
master-9d23589
9d23589d
·
fix pthreads setaffinity usage on android (#2020)
·
Jun 27, 2023
master-0be54f7
0be54f75
·
baby-llama : fix build after ggml_rope change (#2016)
·
Jun 27, 2023
master-eaa6ca5
eaa6ca5a
·
ggml : increase max tensor name + clean up compiler warnings in train-text (#1988)
·
Jun 26, 2023
master-c824d2e
c824d2e3
·
ggml : avoid conv 2d kernel round up
·
Jun 26, 2023
master-b853d45
b853d456
·
ggml : add NUMA support (#1556)
·
Jun 26, 2023
master-9225bae
9225baef
·
k-quants : fix indentation
·
Jun 26, 2023
master-a84ab1d
a84ab1da
·
tests : fix quantize perf (#1990)
·
Jun 26, 2023
master-5743ca8
5743ca80
·
k-quants : add AVX support to dot functions (#1916)
·
Jun 26, 2023
master-6769e94
6769e944
·
k-quants : support for super-block size of 64 (#2001)
·
Jun 26, 2023
master-cbebf61
cbebf61c
·
Fix assert when free invalid cuda pointer (#2005)
·
Jun 26, 2023
master-bd34cdd
bd34cdde
·
ggml : sync latest ggml (custom operators)
·
Jun 25, 2023
master-c2a08f8
c2a08f87
·
fix server sampling: top k sampler first (#1977)
·
Jun 25, 2023
master-5ec8dd5
5ec8dd5a
·
#1869 Fix null reference errors when training from scratch with CUDA (#1907)
·
Jun 24, 2023
master-65bdd52
65bdd52a
·
tests : sync test-grad0 from ggml
·
Jun 24, 2023
master-f2c754e
f2c754e1
·
ggml : improve ggml_graph_dump_dot, add ggml_format_name (#1978)
·
Jun 24, 2023
master-b061ba9
b061ba9e
·
llama : fix top-p sampling to match the canonical definition (#1953)
·
Jun 24, 2023
master-527b6fb
527b6fba
·
llama : make model stateless and context stateful (llama_state) (#1797)
·
Jun 24, 2023
1
…
12
13
14
15
16
17
18
19
20
…
38