 SeungWon Jeong
|
fb215c3832
server : normalize embeddings (#5956)
|
hai 1 ano |
 Minsoo Cheong
|
6d341ab6c5
speculative : implement stochastic speculative sampling (#5625)
|
hai 1 ano |
 DAN™
|
82f3e668ad
common : use LLAMA_DEFAULT_SEED (#5855)
|
hai 1 ano |
 Douglas Hanley
|
475df1d6cf
llama : allow for user specified embedding pooling type (#5849)
|
hai 1 ano |
 Pierrick Hymbert
|
3ab8b3a92e
llama : cleanup unused mmq flags (#5772)
|
hai 1 ano |
 Georgi Gerganov
|
9d533a77d0
llama : fix defrag bugs + add parameter (#5735)
|
hai 1 ano |
|
Georgi Gerganov
|
ab336a9d5e
code : normalize enum names (#5697)
|
hai 1 ano |
 Alexey Parfenov
|
6dcc02d244
server : add "samplers" param to control the samplers order (#5494)
|
hai 1 ano |
 bmwl
|
f486f6e1e5
ggml : add numa options (#5377)
|
hai 1 ano |
|
Alexey Parfenov
|
a803333a4e
common : use enums for sampler types (#5418)
|
hai 1 ano |
 Jared Van Bortel
|
1ec3332ade
YaRN : store rope scaling type as int32_t in memory (#5285)
|
hai 1 ano |
|
Georgi Gerganov
|
5cb04dbc16
llama : remove LLAMA_MAX_DEVICES and LLAMA_SUPPORTS_GPU_OFFLOAD (#5240)
|
%!s(int64=2) %!d(string=hai) anos |
 Kawrakow
|
6f9939d119
KL-divergence (#5076)
|
%!s(int64=2) %!d(string=hai) anos |
|
Kawrakow
|
7dcbe39d36
Add ability to evauate multiple choice tasks (#5047)
|
%!s(int64=2) %!d(string=hai) anos |
|
Kawrakow
|
682986a08e
Add Winogrande evaluation (#5015)
|
%!s(int64=2) %!d(string=hai) anos |
 stduhpf
|
e0324285a5
speculative : threading options (#4959)
|
%!s(int64=2) %!d(string=hai) anos |
 Yann Follet
|
722d33f34e
main : add parameter --no-display-prompt (#4541)
|
%!s(int64=2) %!d(string=hai) anos |
 slaren
|
e7e4df031b
llama : ggml-backend integration (#4766)
|
%!s(int64=2) %!d(string=hai) anos |
|
Georgi Gerganov
|
7edefbd79c
main : better name for variable n_print (#4874)
|
%!s(int64=2) %!d(string=hai) anos |
|
Georgi Gerganov
|
3ca63b4538
main : disable token count by default (#4874)
|
%!s(int64=2) %!d(string=hai) anos |
 pudepiedj
|
43f76bf1c3
main : print total token count and tokens consumed so far (#4874)
|
%!s(int64=2) %!d(string=hai) anos |
|
Georgi Gerganov
|
52531fdff8
main : add self-extend support (#4815)
|
%!s(int64=2) %!d(string=hai) anos |
 LeonEricsson
|
7082d24cec
lookup : add prompt lookup decoding example (#4484)
|
%!s(int64=2) %!d(string=hai) anos |
|
Georgi Gerganov
|
bcc0eb4591
llama : per-layer KV cache + quantum K cache (#4309)
|
%!s(int64=2) %!d(string=hai) anos |
 Kerfuffle
|
5aa365d88f
llama : allow overriding GGUF metadata when loading model (#4092)
|
%!s(int64=2) %!d(string=hai) anos |
 MaggotHATE
|
52c8bc3cf3
sampling : custom samplers order (#4285)
|
%!s(int64=2) %!d(string=hai) anos |
|
Georgi Gerganov
|
6b0a7420d0
llama : KV cache view API + better KV cache management (#4170)
|
%!s(int64=2) %!d(string=hai) anos |
 Seb C
|
881800d1f0
main : Add ChatML functionality to main example (#4046)
|
%!s(int64=2) %!d(string=hai) anos |
|
Kerfuffle
|
91f6499393
Respect tokenizer.ggml.add_bos_token value when tokenizing (#4040)
|
%!s(int64=2) %!d(string=hai) anos |
|
Georgi Gerganov
|
8f961abdc4
speculative : change default p_accept to 0.5 + CLI args (#3919)
|
%!s(int64=2) %!d(string=hai) anos |
