Skip to content
GitLab
Explore
Sign in
Overview
Active
Stale
All
This project is mirrored from
https://github.com/ggerganov/llama.cpp
. Pull mirroring updated
Sep 19, 2024
.
convert-lora-fix
7935986f
·
fix convert-lora-to-ggml.py
·
Aug 23, 2023
fix-eos
977629a3
·
Merge branch 'master' into fix-eos
·
Aug 23, 2023
fix-whitespace
d8beb85c
·
Merge branch 'master' into fix-whitespace
·
Aug 23, 2023
fix-falcon-cuda
ac4bb6ba
·
cuda : add RoPE kernel for mode == 2 (NeoX)
·
Aug 24, 2023
metal-add-q8_0
1202e06c
·
metal : add Q8_0 mul_mm kernel
·
Aug 24, 2023
codellama-gguf-rope-base
21dcd944
·
gguf : add rope_freq_base parameter for CodeLlama
·
Aug 24, 2023
metal-fix-memory-leak
53dea117
·
metal : fix encoders memory leak
·
Aug 24, 2023
codellama-ctx
75945403
·
convert.py : try to determine n_ctx automatically for CodeLlama
·
Aug 24, 2023
codellama-hf-freq-base
06f79259
·
convert.py : add freq_base when converting CodeLlama from an HF model
·
Aug 24, 2023
gguf-pip
0248ca81
·
gguf : add notes for tests
·
Aug 25, 2023
gguf-export-objs
058fbdd8
·
gguf : bump version
·
Aug 25, 2023
ci-gguf
e38f7847
·
ci : pip install gguf in editable mode
·
Aug 25, 2023
llama-bench-model-size
bc0dc16c
·
more compact markdown output
·
Aug 25, 2023
ik/faster_ppl
ce45974a
·
Faster perplexity computation
·
Aug 25, 2023
ik/fix_hellaswag
d34472c1
·
Fix HellaSwag
·
Aug 26, 2023
fix-code-llama-quantum-mixtures
eff86d4f
·
k-quants : remove unnecessary tensor shape restrictions
·
Aug 26, 2023
ik/refine_70B
3979af1e
·
PR comment
·
Aug 26, 2023
ik/fix_cuda_qkk64
f547c585
·
Make ggml-cuda.cu build with QK_K = 64
·
Aug 26, 2023
fix-tokenizer
724fa67d
·
perplexity.cpp : better way to deal wirh spm prepending space
·
Aug 26, 2023
gguf-64bit
33a5517d
·
llama.cpp : print gguf version
·
Aug 26, 2023
Prev
1
2
3
4
5
6
7
8
9
10
…
26
Next