klosax
4cef57c81a
convert-llama-h5-to-gguf.py : no need to convert tensors twice
2023-08-12 21:50:24 +02:00
klosax
8f09157ec9
convert-gptneox-h5-to-gguf.py : no need to convert tensors twice
2023-08-12 21:48:58 +02:00
klosax
5d81a715d4
gguf.py : no need to convert tensors twice
2023-08-12 21:45:45 +02:00
M. Yusuf Sarıgöz
60d540831b
gguf : roper closing of file
2023-08-12 21:42:31 +03:00
M. Yusuf Sarıgöz
202eab04d3
gguf : quantization is working
2023-08-12 16:39:05 +03:00
M. Yusuf Sarıgöz
1fc3d30b71
gguf : start implementing quantization (WIP)
2023-08-12 16:09:47 +03:00
M. Yusuf Sarıgöz
fa7c39540c
gguf : start implementing quantization (WIP)
2023-08-12 15:55:58 +03:00
M. Yusuf Sarıgöz
b2571af255
gguf : start implementing quantization (WIP)
2023-08-12 14:28:17 +03:00
M. Yusuf Sarıgöz
c4f02b4f74
gguf : start implementing quantization (WIP)
2023-08-12 12:01:17 +03:00
M. Yusuf Sarıgöz
0e1a3c7e7d
gguf : start implementing quantization (WIP)
2023-08-12 11:32:34 +03:00
M. Yusuf Sarıgöz
4fa017a1f9
gguf : start implementing quantization (WIP)
2023-08-12 10:40:56 +03:00
M. Yusuf Sarıgöz
186c496fdf
Merge branch 'gguf' of https://github.com//ggerganov/llama.cpp into gguf
2023-08-12 07:25:10 +03:00
M. Yusuf Sarıgöz
2f52008b20
gguf : rm references to old file magics
2023-08-12 07:24:46 +03:00
byte-6174
b19edd54d5
Adding support for llama2.c models ( #2559 )
2023-08-12 01:17:25 +02:00
Equim
53dc399472
server: fixed wrong variable name in timing json ( #2579 )
...
* server: fixed wrong variable name in timing json
* remove redunct entry
2023-08-12 00:35:14 +02:00
klosax
e76c59d524
Update gptneox-main.cpp
2023-08-11 23:09:49 +02:00
klosax
2a5ac7af44
Update gguf_tensor_map.py
2023-08-11 23:08:48 +02:00
M. Yusuf Sarıgöz
e732423280
gguf : get rid of n_mult, read n_ff from file
2023-08-11 23:50:38 +03:00
M. Yusuf Sarıgöz
f44bbd3d88
gguf : rm redundant method
2023-08-11 21:00:51 +03:00
M. Yusuf Sarıgöz
7009cf581c
gguf : shorter name for member variable
2023-08-11 20:43:02 +03:00
M. Yusuf Sarıgöz
61919c1a8f
gguf : rm references to old file formats
2023-08-11 20:36:11 +03:00
M. Yusuf Sarıgöz
d09fd10713
gguf : write metadata in gguf_file_saver
2023-08-11 20:07:43 +03:00
M. Yusuf Sarıgöz
781b9ec3f5
gguf : write metadata in gguf_file_saver (WIP)
2023-08-11 18:01:26 +03:00
M. Yusuf Sarıgöz
28abfc90fa
gguf : write metadata in gguf_file_saver (WIP)
2023-08-11 13:27:58 +03:00
M. Yusuf Sarıgöz
e3a4960953
gguf : add gguf_get_kv_type
2023-08-11 13:03:23 +03:00
M. Yusuf Sarıgöz
eb8ca6996f
gguf : add gguf_get_kv_type
2023-08-11 12:24:08 +03:00
M. Yusuf Sarıgöz
b2440f1943
gguf : start implementing gguf_file_saver (WIP)
2023-08-11 11:29:50 +03:00
M. Yusuf Sarıgöz
a356b0e228
gguf : start implementing gguf_file_saver (WIP)
2023-08-11 10:50:02 +03:00
M. Yusuf Sarıgöz
e7d346c37c
gguf : start implementing gguf_file_saver (WIP)
2023-08-11 09:52:01 +03:00
DannyDaemonic
9ca4abed89
Handle ENABLE_VIRTUAL_TERMINAL_PROCESSING more gracefully on earlier versions of Windows.
2023-08-10 13:11:36 -07:00
M. Yusuf Sarıgöz
f316b94c7c
gguf : rm deprecated function
2023-08-10 20:20:22 +03:00
M. Yusuf Sarıgöz
cfb8e35b73
gguf : inference with 7B model working (WIP)
2023-08-10 19:56:56 +03:00
M. Yusuf Sarıgöz
42cc04d11d
gguf : calculate n_mult
2023-08-10 18:49:08 +03:00
M. Yusuf Sarıgöz
22de6c5c4c
upd .gitignore
2023-08-10 18:09:49 +03:00
M. Yusuf Sarıgöz
4c0f64e302
rm binary commited by mistake
2023-08-10 18:07:41 +03:00
M. Yusuf Sarıgöz
4f865181aa
gguf : start implementing libllama in GGUF (WIP)
2023-08-10 17:49:31 +03:00
Christian Demsar
e59fcb2bc1
Add --n-predict -2 for stopping generation on full context ( #2565 )
2023-08-10 16:28:27 +02:00
M. Yusuf Sarıgöz
1c4d8bf981
gguf : start implementing libllama in GGUF (WIP)
2023-08-10 16:52:08 +03:00
Martin Krasser
1638757767
Fix grammar-based sampling issue in server ( #2566 )
2023-08-10 13:16:38 +03:00
Sam Spilsbury
916a9acdd0
ggml-alloc: Don't try to re-use buffers of external tensors ( #2562 )
...
* ggml-alloc: Don't try to re-use buffers of external tensors
They might be weights that came from another context, so we
have no control over them (and they might be re-used elsewhere
so writing to them would be a bad idea).
* ggml-alloc: >= when checking for out-of-bounds
Co-authored-by: slaren <slarengh@gmail.com >
---------
Co-authored-by: slaren <slarengh@gmail.com >
2023-08-09 22:47:42 +02:00
grahameth
ea04a4ca19
add log_callback to llama_context_params for custom logging. ( #2234 )
...
* add log_callback to llama_context_params for custom logging.
* Fix macro expansion on gcc
* Add struct llama_state for global variables and move log_callback there
* Turn log level into enum and some minor changes.
* Remove model_for_logging parameter (not needed anymore)
* Convert remaining fprintf(stderr, ...) calls to use new macros.
* Fix enum and initialize g_state
* Fix log calls after merge
* Fix missing static
* Add back all the new lines in the logging strings
* Add comment for llama_log_callback and replace remaining printf calls
---------
Co-authored-by: grahameth <->
Co-authored-by: Helmut <helmut.buhler@inf.h-brs.de>
2023-08-09 22:46:40 +02:00
Johannes Gäßler
25d43e0eb5
CUDA: tuned mul_mat_q kernels ( #2546 )
2023-08-09 09:42:34 +02:00
klosax
0246d0dd6f
gptneox-main.cpp : map tensor names
2023-08-09 00:54:21 +02:00
klosax
7d5f4522dd
convert-llama-h5-to-gguf.py : map tensor names
2023-08-09 00:52:16 +02:00
klosax
f4d137d98c
convert-gptneox-h5-to-gguf.py : map tensor names
2023-08-09 00:50:11 +02:00
klosax
ece4fc185e
map tensor names
2023-08-09 00:48:33 +02:00
Martin Krasser
f5bfea0580
Allow passing grammar to completion endpoint ( #2532 )
...
* Allow passing grammar to completion endpoint
2023-08-08 16:29:19 +03:00
Johannes Gäßler
acfc5478ff
CUDA: tighter VRAM scratch size for 65b/70b ( #2551 )
2023-08-08 14:38:16 +02:00
chaihahaha
7ed8d1fe7f
llm.vim : multiline autocompletion, get rid of "^@" ( #2543 )
2023-08-08 15:07:02 +03:00
Georgi Gerganov
e7f94d6fdc
vim : bring back simple llm.vim example
2023-08-08 15:06:18 +03:00
