ggml-org/llama.cpp

Pull Requests Commits
Revert "finetune : add --n-gpu-layers flag info to --help (#4128)"
ggerganov committed 2 years ago
Verified dae06c06
finetune : add --n-gpu-layers flag info to --help (#4128)
csaben committed 2 years ago
Verified 05e8301e
server : relay error messages (#4131)
SoftwareRenderer committed 2 years ago
Verified 936c79b2
common : comma should be semicolon (#4137)
kchro3 committed 2 years ago
Verified 262005ad
gitignore : tokenize
ggerganov committed 2 years ago
Verified 35985acf
gguf-py : export chat templates (#4125)
slaren committed 2 years ago
Verified e9370664
tokenize example: Respect normal add BOS token behavior (#4126)
KerfuffleV2 committed 2 years ago
Verified 28a2e6e7
scripts : Remove missed baichuan convert script (#4127)
Galunid committed 2 years ago
Verified 0b5c3b04
Clean up ggml-cuda.cu warnings when compiling with clang (for ROCM) (#4124)
KerfuffleV2 committed 2 years ago
Verified 2923f17f
llama : increase max nodes (#4115)
slaren committed 2 years ago
Verified bbecf3f4
build : support ppc64le build for make and CMake (#3963)
bufferoverflow committed 2 years ago
Verified 8e936108
tokenize : fix trailing whitespace
ggerganov committed 2 years ago
Verified 5ad387e9
examples : add tokenize (#4039)
zakkor committed 2 years ago
Verified 2fa02b4b
convert : use 'model' value if it exists. This allows karpathy/tinyllamas to load (#4089)
dmahurin committed 2 years ago
Verified 2ab0707a
py : Falcon HF compatibility (#4104)
cmp-nct committed 2 years ago
Verified 11173c92
common : improve yaml log escaping (#4080)
joennlae committed 2 years ago
Verified 9e87ef60
llava : fix compilation warning that fread return value is not used (#4069)
huawei-lin committed 2 years ago
Verified c7cce124
py : remove superfluous import statements (#4076)
jpodivin committed 2 years ago
Verified f7d5e975
train : move number of gpu layers argument parsing to common/train.cpp (#4074)
jpodivin committed 2 years ago
Verified ba4cf5c0
llama : add functions to get the model's metadata (#4013)
slaren committed 2 years ago
Verified e85bb1a8
finetune : speed-up ggml_compute_forward_out_prod_f32 via BLAS (#4079)
gwjr committed 2 years ago
Verified 3e916a07
finetune : zero the loraB initial vectors (#4082)
AndrewGodfrey committed 2 years ago
Verified 947f64f1
cuda : get_row_rounding F32 (#4095)
AndrewGodfrey committed 2 years ago
Verified b83e149e
llama : fix data units (#4101)
ggerganov committed 2 years ago
Verified 4f447a48
Respect tokenizer.ggml.add_bos_token value when tokenizing (#4040)
KerfuffleV2 committed 2 years ago
Verified 91f64993
gguf : fix potential infinite loops while parsing (#4100)
texmex76 committed 2 years ago
Verified 8da46278
llama : restore prefix space in llama tokenizer (#4081)
cebtenzzre committed 2 years ago
Verified a6fc554e
ggml-cuda : increase max graph size (#4084)
slaren committed 2 years ago
Verified 1cf2850d
Fix MacOS Sonoma model quantization (#4052)
TortoiseHam committed 2 years ago
Verified 6bb4908a
stablelm : StableLM support (#3586)
Galunid committed 2 years ago
Verified 36eed0c4
Terms Of Use Privacy Policy Refund Policy Impressum