ggml-org/llama.cpp

Pull Requests Commits

gguf.py : indentation

ggerganov committed 2 years ago

Verified 6a9e6375

Merge branch 'gguf' into gguf-write-single-pass

ggerganov committed 2 years ago

Verified 307e09cd

gguf.py : fix vertical alignment

ggerganov committed 2 years ago

Verified e426b3cf

llama : fix tensor name grepping during quantization

ggerganov committed 2 years ago

Verified 5484737d

llama : throw error if gguf fails to init from file

ggerganov committed 2 years ago

Verified 57eaadb8

llama.cpp : typo

klosax committed 2 years ago

Verified b3cc1829

convert.py : fix HF tensor permuting / unpacking

ggerganov committed 2 years ago

Verified acaa9823

quantize-stats.cpp : .bin --> .gguf

klosax committed 2 years ago

Verified 78e1e578

common.h : .bin --> .gguf

klosax committed 2 years ago

Verified fb11dd3f

ggml : fix bug in gguf_set_kv

ggerganov committed 2 years ago

Verified e72c8c21

gguf : rm file_type key and method

monatis committed 2 years ago

4dbce7d0

gguf : refactor pth to gguf conversion script

monatis committed 2 years ago

1d93d04c

llama : fix lambda capture

ggerganov committed 2 years ago

Verified 899f9a53

gptneox : move as a WIP example

ggerganov committed 2 years ago

Verified 93f285bd

gguf : rename h5 to hf (for HuggingFace)

monatis committed 2 years ago

f7170417

llama : fix llama_model_loader memory leak

ggerganov committed 2 years ago

Verified 81a2c2a6

gguf : refactor gptneox conversion script

monatis committed 2 years ago

9f02694c

ci : update ".bin" to ".gguf" extension

ggerganov committed 2 years ago

Verified dd9e2fc9

editorconfig : ignore models folder

ggerganov committed 2 years ago

Verified c3b73937

gguf : style fixes in simple conversion script

monatis committed 2 years ago

22c61c5b

Merge branch 'master' into gguf

ggerganov committed 2 years ago

Verified 6d66ef96

llama : sync with recent PRs on master

ggerganov committed 2 years ago

Verified 11bf4366

gguf : fix conflicts

monatis committed 2 years ago

2f8fc92d

convert.py : better always have n_head_kv and default it to n_head

ggerganov committed 2 years ago

Verified 8ace03ad

convert.py : n_head_kv optional and .gguf file extension

klosax committed 2 years ago

Verified d646c4ef

Revert "ci : disable CI temporary to not waste energy"

ggerganov committed 2 years ago

Verified dd016cc2

convert.py : update to support GGUF output

ggerganov committed 2 years ago

Verified 2ddd9681

convert-new.py : output gguf (#2635)

ggerganov committed 2 years ago

Verified e0429d38

gguf : single pass for writing tensors + refactoring writer

monatis committed 2 years ago

5f97a48f

gguf : single pass for writing tensors + refactoring writer

monatis committed 2 years ago

dce07c31

Older