 jaime-m-p
|
43248e5594
llama3 custom regex split (#6965)
|
пре 1 година |
 Johannes Gäßler
|
a743d76a01
CUDA: generalize FP16 fattn vec kernel (#7061)
|
пре 1 година |
 Daniel Bevenius
|
fd9f92b154
llama : update llama_timings.n_p_eval setting (#7160)
|
пре 1 година |
 Ren Xuancheng
|
229ffff872
llama : add BPE pre-tokenization for Qwen2 (#7114)
|
пре 1 година |
 DAN™
|
4cd621c26d
convert : add BPE pre-tokenization for DBRX (#7132)
|
пре 1 година |
 Justine Tunney
|
3855416027
ggml : introduce bfloat16 support (#6412)
|
пре 1 година |
 nopperl
|
b6aa670203
Fix OLMo HF to GGUF conversion (#6910)
|
пре 1 година |
|
DAN™
|
889bdd7686
command-r : add BPE pre-tokenization (#7063)
|
пре 1 година |
 Georgi Gerganov
|
92139b90af
tests : add test-tokenizer-0.sh + fix some tokenizers (#7036)
|
пре 1 година |
 alwqx
|
6ecf3189e0
chore: fix typo in llama.cpp (#7032)
|
пре 1 година |
|
Georgi Gerganov
|
9c67c2773d
ggml : add Flash Attention (#5021)
|
пре 1 година |
|
Georgi Gerganov
|
f4ab2a4147
llama : fix BPE pre-tokenization (#6920)
|
пре 1 година |
|
Johannes Gäßler
|
c4f708a93f
llama : fix typo LAMMAFILE -> LLAMAFILE (#6974)
|
пре 1 година |
 Xuan Son Nguyen
|
7bb36ccf91
gguf : enforce that tensor names are unique (#6905)
|
пре 1 година |
 agray3
|
928e0b7013
Reset schedule earlier to allow overlap with ggml graph computation on device (#6933)
|
пре 1 година |
 Pierrick Hymbert
|
0c4d489e29
quantize: add imatrix and dataset metadata in GGUF (#6658)
|
пре 1 година |
 slaren
|
017e6999b5
add basic tensor data validation function (#6884)
|
пре 1 година |
|
Georgi Gerganov
|
dba497e0c1
cmake : restore LLAMA_LLAMAFILE_DEFAULT
|
пре 1 година |
|
slaren
|
d6e1d44f16
llama : synchronize before get/set session data (#6911)
|
пре 1 година |
|
slaren
|
0ead1f1072
llama : check that all the tensor data is in the model file (#6885)
|
пре 1 година |
|
Georgi Gerganov
|
aa750c1ede
tests : minor bash stuff (#6902)
|
пре 1 година |
 jiez
|
1966eb2615
quantize : add '--keep-split' to quantize model into shards (#6688)
|
пре 1 година |
 Douglas Hanley
|
b4e4b8a935
llama : add llama_get_pooling_type function (#6862)
|
пре 1 година |
|
Johannes Gäßler
|
28103f4832
Server: fix seed for multiple slots (#6835)
|
пре 1 година |
 Tristan Druyen
|
abd3314064
llama : add phi 3 chat template (#6857)
|
пре 1 година |
 liuwei-git
|
c8297c6af5
llama : add phi3 support (#6852)
|
пре 1 година |
|
Georgi Gerganov
|
8960fe86ae
llama : fix typo in <|im_end|> token text (#6745)
|
пре 1 година |
|
Georgi Gerganov
|
40f74e4d73
llama : add option to render special/control tokens (#6807)
|
пре 1 година |
 Wouter
|
7dbdba5690
llama : add llama-3 chat template (#6751)
|
пре 1 година |
 Pedro Cuenca
|
b97bc3966e
llama : support Llama 3 HF conversion (#6745)
|
пре 1 година |
