vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
712 commits 380 branches 3056 tags 365 MiB
Commit graph

712 commits

This branch
This branch
All branches
Author SHA1 Message Date
Georgi Gerganov
f9be42add0
readme : add quantization info 2023-04-26 23:24:42 +03:00
Georgi Gerganov
574406dc7e
ggml : add Q5_0 and Q5_1 quantization (#1187) 
* ggml : add Q5_0 quantization (cuBLAS only)

* ggml : fix Q5_0 qh -> uint32_t

* ggml : fix q5_0 histogram stats

* ggml : q5_0 scalar dot product

* ggml : q5_0 ARM NEON dot

* ggml : q5_0 more efficient ARM NEON using uint64_t masks

* ggml : rename Q5_0 -> Q5_1

* ggml : adding Q5_0 mode

* quantize : add Q5_0 and Q5_1 to map

* ggml : AVX2 optimizations for Q5_0, Q5_1 (#1195)

---------

Co-authored-by: Stephan Walter <stephan@walter.name>
 2023-04-26 23:14:13 +03:00
Ásgeir Bjarni Ingvarsson
87a6f846d3
Allow setting the rng seed after initialization. (#1184) 
The llama_set_state_data function restores the rng state to what it
was at the time llama_copy_state_data was called. But users may want
to restore the state and proceed with a different seed.
 2023-04-26 22:08:43 +02:00
DaniAndTheWeb
ea3ad7eb60
Updating build instructions to include BLAS support (#1183) 
* Updated build information

First update to the build instructions to include BLAS.

* Update README.md

* Update information about BLAS

* Better BLAS explanation

Adding a clearer BLAS explanation and adding a link to download the CUDA toolkit.

* Better BLAS explanation

* BLAS for Mac

Specifying that BLAS is already supported on Macs using the Accelerate Framework.

* Clarify the effect of BLAS

* Windows Make instructions

Added the instructions to build with Make on Windows

* Fixing typo

* Fix trailing whitespace
 2023-04-26 22:03:03 +02:00
Pavol Rusnak
859fee6dfb
quantize : use map to assign quantization type from string (#1191) 
instead of `int` (while `int` option still being supported)

This allows the following usage:

`./quantize ggml-model-f16.bin ggml-model-q4_0.bin q4_0`

instead of:

`./quantize ggml-model-f16.bin ggml-model-q4_0.bin 2`
 2023-04-26 18:43:27 +02:00
Concedo
101f7a6e73 updated readme 2023-04-26 23:50:00 +08:00
Concedo
93a8e00dfa Merge branch 'master' into concedo 
# Conflicts:
#	flake.nix
 2023-04-26 18:01:35 +08:00
Disty0
27bc29128e
Update README.md (#120) 2023-04-26 17:33:34 +08:00
Stephan Walter
4afcc37869
Update SHA256SUMS after quantization change (#1181) 
Co-authored-by: Pavol Rusnak <pavol@rusnak.io>
 2023-04-25 23:41:56 +02:00
ostix360
667c501334
py : cast lora_alpha to int in convert-lora-to-ggml (#1170) 
Co-authored-by: Pavol Rusnak <pavol@rusnak.io>
 2023-04-25 23:33:08 +02:00
Pavol Rusnak
bb98e77be7
nix: use convert.py instead of legacy wrapper convert-pth-to-ggml.py (#981) 2023-04-25 23:19:57 +02:00
Georgi Gerganov
7a32fcb3b2
ggml : add Q8_0 quantization format (rename the old one to Q8_1) (ARM NEON) (#1179) 
* ggml : add Q8_0 quantization format (rename the old one to Q8_1)

* tests : fix test-quantize-fns

* ggml : finalize Q8_0 implementation

* ggml : use q4_0_q8_0 and q4_2_q8_0

* ggml : fix Q8_0 dot product bug (ARM)

* ggml : Q8_0 unroll x2

* ggml : fix bug - using wrong block type

* ggml : extend quantize_fns_t with "vec_dot_type"

* ggml : fix Q8_0 to use 255 values out of 256

* ggml : fix assert using wrong QK4_2 instead of QK4_3
 2023-04-25 23:40:51 +03:00
unbounded
dd0eabc049
ggml : use full range for Q4_0 and Q4_2 quantization (#729) 
* Use full range for q4_0 quantization

By keeping the sign of the highest magnitude, we can make sure the
highest value maps to -8, which is currently unused.
This is a bit of a freebie since it is fully backwards compatible with
the current format.

* Update quantize_row_q4_0 for AVX/AVX2

* Update quantize_row_q4_0 for WASM

Untested

* Update quantize_row_q4_0 for Arm NEON

* Update quantize_row_q4_0 for PowerPC

Untested

* Use full range for q4_2 quantization
 2023-04-25 20:20:46 +03:00
Concedo
0aa3d839fb free old ctx on retry 2023-04-25 23:42:57 +08:00
Concedo
a696b0a16c missed another thing 2023-04-25 23:16:04 +08:00
Concedo
8c9c218609 missed a thing 2023-04-25 23:02:08 +08:00
Concedo
235daf4016 Merge branch 'master' into concedo 
# Conflicts:
#	.github/workflows/build.yml
#	README.md
 2023-04-25 20:44:22 +08:00
Concedo
72b2331ad6 edge cases with mem crash? need verify 2023-04-25 20:42:30 +08:00
Concedo
5eec5d6ed9 Added backwards compatibility to an earlier version of NeoX. 2023-04-25 20:34:18 +08:00
Concedo
bff998f871 Slight refactor of the python code: credits to @LuxF3rre 2023-04-25 19:20:14 +08:00
xaedes
54bb60e268
ggml : fix bug in ggml_compute_forward_sum_f32 (#1162) 
The sum over all rows is now computed instead of just the last row
 2023-04-24 23:02:02 +02:00
Georgi Gerganov
8a0f8673ba
ggml : export symbols (#1155) 2023-04-24 22:18:25 +03:00
xaedes
0c5692345d
examples : add save_load_state example (#1150) 
* add save_load_state example

* use <cstdio> instead of <iostream> and fprintf / printf instead of cout

* renamed save-load-state example files replacing underscores by dashes
 2023-04-24 19:23:31 +03:00
Georgi Gerganov
957c8ae21d
llama : increase scratch buffer size for 65B (ref #1152) 
Temporary solution
 2023-04-24 18:47:30 +03:00
mgroeber9110
9b0a4d4214
examples/main README improvements and some light refactoring (#1131) 2023-04-24 15:45:32 +00:00
Stephan Walter
2ec83428de
Fix build for gcc 8 and test in CI (#1154) 2023-04-24 15:38:26 +00:00
slaren
e4cf982e0d
Fix cuda compilation (#1128) 
* Fix: Issue with CUBLAS compilation error due to missing -fPIC flag

---------

Co-authored-by: B1gM8c <89020353+B1gM8c@users.noreply.github.com>
 2023-04-24 17:29:58 +02:00
Concedo
59fb174678 fixed compile errors, made mmap automatic when lora is selected, added updated quantizers and quantization handling for gpt neox gpt 2 and gptj 2023-04-24 23:20:06 +08:00
Concedo
3962eb39c7 added token unbanning 2023-04-24 21:50:20 +08:00
Concedo
1b9b9068b1 merged q4_2 and q4_3 dequants and FIXED CLBLAST SLOWNESS! 2023-04-24 21:33:01 +08:00
Concedo
e58f1d1336 Merge branch 'master' into concedo_experimental 2023-04-24 19:43:17 +08:00
Georgi Gerganov
c4fe84fb0d
llama : refactor get / set state + remove redundant kv cache API (#1143) 2023-04-24 07:40:02 +03:00
Concedo
8e615c8245 Merge branch 'master' into concedo_experimental 
# Conflicts:
#	README.md
 2023-04-24 12:20:08 +08:00
slaren
1d78fecdab
Fix LoRA acronym (#1145) 2023-04-23 23:03:44 +02:00
Georgi Gerganov
284685f169
scripts : add helper scripts to synch ggml repo 2023-04-23 19:57:09 +03:00
DannyDaemonic
edce63baa9
Added README.md for main with examples and explanations (#1139) 2023-04-23 15:37:02 +00:00
Georgi Gerganov
ec9cdb6752
ggml : do not print perf ops that have not been used at all 2023-04-23 18:32:52 +03:00
Georgi Gerganov
e4422e299c
ggml : better PERF prints + support "LLAMA_PERF=1 make" 2023-04-23 18:15:39 +03:00
Stephan Walter
53c8434398
Improve AVX2 for vec_dot_q4_3_q8_0 (#1138) 2023-04-23 11:01:03 +00:00
Pavol Rusnak
c6524f46eb
readme : update gpt4all instructions (#980) 2023-04-23 10:21:26 +02:00
Concedo
9129e937f9 only llama can use batch sizes above 256 to prevent unacceptably high memory usage 2023-04-23 15:57:06 +08:00
Yishuo Wang
c9e2c26f41
A better packNibbles and mul_sum_i8_pairs_float implementation using AVX512 (#1119) 2023-04-23 07:57:05 +00:00
Concedo
432cc91649 still needs to be a bit higher for very small contexts 2023-04-23 15:01:38 +08:00
Concedo
4e1ea2ac61 hopefully fixed the ooms for good 2023-04-23 13:49:50 +08:00
Gustavo Rocha Dias
3f21bd81f3
doc - Better explanation of how to build the libraries at Windows. (#107) 2023-04-23 13:40:09 +08:00
Concedo
d41490c27b just revert back to the working commit 2023-04-23 00:35:42 +08:00
Concedo
c60fb5ef4b fixed rwkv build errors on ARM devices 2023-04-23 00:18:38 +08:00
Concedo
b5d6284190 increase initial buffer too 2023-04-23 00:07:33 +08:00
Concedo
d2f14b2b1f add an extra buffer to mem allocations 2023-04-23 00:04:32 +08:00
Concedo
7c60441d71 Merge branch 'master' into concedo 
# Conflicts:
#	.github/workflows/build.yml
#	CMakeLists.txt
 2023-04-22 23:46:14 +08:00