 Cebtenzzre
|
bc39553c90
build : enable more non-default compiler warnings (#3200)
|
2 жил өмнө |
 slaren
|
16bc66d947
llama.cpp : split llama_context_params into model and context params (#3301)
|
2 жил өмнө |
 xaedes
|
0e76a8992c
train : finetune LORA (#2632)
|
2 жил өмнө |
|
Cebtenzzre
|
ecf90b1a51
gguf : make token scores and types optional (#3347)
|
2 жил өмнө |
 Georgi Gerganov
|
ec893798b7
llama : custom attention mask + parallel decoding + no context swaps (#3228)
|
2 жил өмнө |
|
Cebtenzzre
|
20c7e1e804
gguf : fix a few general keys (#3341)
|
2 жил өмнө |
 Rickard Hallerbäck
|
dc6897404e
metal : reusing llama.cpp logging (#3152)
|
2 жил өмнө |
 Johannes Gäßler
|
8185710a80
CUDA: use only 1 thread if fully offloaded (#2915)
|
2 жил өмнө |
|
Cebtenzzre
|
a5661d7e71
llama : allow gguf RoPE keys to be overridden with defaults (#3240)
|
2 жил өмнө |
|
slaren
|
8b428c9bc8
llama.cpp : show model size and BPW on load (#3223)
|
2 жил өмнө |
 goerch
|
b08e75baea
Fixing the last deviations from sentencepiece indicated by test-tokenizer-1 (#3170)
|
2 жил өмнө |
|
Cebtenzzre
|
3aefaab9e5
check C++ code with -Wmissing-declarations (#3184)
|
2 жил өмнө |
 Meng Zhang
|
4fe09dfe66
llama : add support for StarCoder model architectures (#3187)
|
2 жил өмнө |
|
Georgi Gerganov
|
a51b687657
metal : relax conditions on fast matrix multiplication kernel (#3168)
|
2 жил өмнө |
|
Cebtenzzre
|
98311c4277
llama : make quantize example up to 2.7x faster (#3115)
|
2 жил өмнө |
 jameswu2014
|
4c8643dd6e
feature : support Baichuan serial models (#3009)
|
2 жил өмнө |
|
goerch
|
71ca2fad7d
whisper : tokenizer fix + re-enable tokenizer test for LLaMa (#3096)
|
2 жил өмнө |
|
Cebtenzzre
|
e64f5b5578
examples : make n_ctx warning work again (#3066)
|
2 жил өмнө |
 Przemysław Pawełczyk
|
cb6c44c5e0
build : do not use _GNU_SOURCE gratuitously (#2035)
|
2 жил өмнө |
 Kunshang Ji
|
7f412dab9c
enable CPU HBM (#2603)
|
2 жил өмнө |
|
Cebtenzzre
|
00d62adb79
fix some warnings from gcc and clang-tidy (#3038)
|
2 жил өмнө |
|
Przemysław Pawełczyk
|
fec2fb19e4
ggml : posixify madvise and pagesize (#3037)
|
2 жил өмнө |
|
Georgi Gerganov
|
35938ee3b0
llama : update logic for number of threads when using BLAS
|
2 жил өмнө |
|
Georgi Gerganov
|
921772104b
speculative : add grammar support (#2991)
|
2 жил өмнө |
|
Georgi Gerganov
|
e36ecdccc8
build : on Mac OS enable Metal by default (#2901)
|
2 жил өмнө |
 opparco
|
3730134776
llama : fix bpe tokenize from byte (#2889)
|
2 жил өмнө |
 momonga
|
c42f0ec6b3
examples : fix gpt-neox (#2943)
|
2 жил өмнө |
 Kerfuffle
|
5d6f19f16b
Allow quantize to only copy tensors, some other improvements (#2931)
|
2 жил өмнө |
 m3ndax
|
ee8654bcd0
minor : add const qualifiers (#2853)
|
2 жил өмнө |
|
Cebtenzzre
|
ef15649972
build : fix most gcc and clang warnings (#2861)
|
2 жил өмнө |
