ngxson/llama.cpp

Pull Requests Commits

correct llama2 chat template

ngxson committed 1 year ago

8b2234ee

cli: add EOT when user hit Ctrl+C (#8296)

ngxson committed 1 year ago

Verified a38b884c

llama : add OpenELM support (#7359)

icecream95 committed 1 year ago

Verified d7fd29ff

tokenize : add --show-count (token) option (#8299)

danbev committed 1 year ago

Verified 6f63d646

build: Export hf-to-gguf as snakecase

ditsuke committed 2 years ago

51d2ebad

doc: Add context for why we add an explicit pytorch source

ditsuke committed 2 years ago

1e920018

chore: Remove rebase artifacts

ditsuke committed 2 years ago

01a5f065

chore: Fixup requirements and build

ditsuke committed 2 years ago

07786a61

chore: ignore all __pychache__

ditsuke committed 2 years ago

de14e2ea

fix: Update script paths in CI scripts

ditsuke committed 2 years ago

82192291

fix: Actually include scripts in build

ditsuke committed 2 years ago

b1c3f26e

build(python): Package scripts with pip-0517 compliance

ditsuke committed 2 years ago

b0a46993

Inference support for T5 and FLAN-T5 model families (#5763)

fairydreaming committed 2 years ago

Verified 807b0c49

tests : add _CRT_SECURE_NO_WARNINGS for WIN32 (#8231)

danbev committed 2 years ago

Verified f8c4c073

llama : suppress unref var in Windows MSVC (#8150)

danbev committed 2 years ago

Verified 402d6fef

convert : fix gemma v1 tokenizer convert (#8248)

ggerganov committed 2 years ago

Verified 20fc3804

[SYCL] Remove unneeded semicolons (#8280)

AidanBeltonS committed 2 years ago

Verified f6190247

Define and optimize RDNA1 (#8085)

daniandtheweb committed 2 years ago

Verified d23287f1

ppl : fix n_seq_max for perplexity (#8277)

slaren committed 2 years ago

Verified 5f2d4e60

fix phi 3 conversion (#8262)

ngxson committed 2 years ago

Verified 916248af

fix typo (#8267)

foldl committed 2 years ago

Verified f8d6a238

Dequant improvements rebase (#8255)

AidanBeltonS committed 2 years ago

Verified fadde671

fix: add missing short command line argument -mli for multiline-input (#8261)

MistApproach committed 2 years ago

Verified a27152b6

Adding step to `clean` target to remove legacy binary names to reduce upgrade / migration confusion arising from #7809. (#8257)

HanClinto committed 2 years ago

Verified 3e2618bc

Removes multiple newlines at the end of files that is breaking the editorconfig step of CI. (#8258)

HanClinto committed 2 years ago

Verified 07a3fc06

Add `JAIS` model(s) (#8118)

fmz committed 2 years ago

Verified 96896737

convert-hf : print output file name when completed (#8181)

danbev committed 2 years ago

Verified 023b8807

cuda : update supports_op for matrix multiplication (#8245)

slaren committed 2 years ago

Verified 0e0590ad

[SYCL] Fix win build conflict of math library (#8230)

luoyu-intel committed 2 years ago

Verified a9f3b102

[SYCL] Fix the sub group size of Intel (#8106)

luoyu-intel committed 2 years ago

Verified d08c20ed

Older