ed
/
rwkv.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
335 Commits 1 Branch 0 Tags 9 MiB
Commit Graph

335 Commits

Author SHA1 Message Date
saharNooby 3535476987 Update README.md: include info about pre-compiled library 2023-04-03 09:48:53 +04:00
saharNooby 5b2830ed30 Increase memory for overhead from 32 MB to 256 MB 2023-04-03 09:32:58 +04:00
hypnopump a64aaa81ec
initial addition 2023-04-03 00:52:26 +02:00
saharNooby d62a050144 Remove hardcoded memory requirements table 2023-04-02 18:37:45 +04:00
saharNooby 1262ad0456 Fix build errors and warnings 2023-04-02 17:23:39 +04:00
saharNooby f2b1dad22b Add GitHub workflows file 2023-04-02 16:56:04 +04:00
saharNooby 6b4ebc328a Update README.md 2023-04-02 15:28:34 +04:00
saharNooby e0684e8104 Add text generation and chat scripts 2023-04-02 15:03:31 +04:00
saharNooby ee46ad208e Add quantization test back, run ggml tests on first context init 2023-04-02 13:05:17 +04:00
saharNooby 1ecbad3a65 Remove unused files 2023-04-02 12:53:41 +04:00
saharNooby 935d16f5db Move library wrapper to separate file, refactor code 2023-04-02 12:24:40 +04:00
saharNooby 38f9d02d52 Fix quantization from FP16 2023-04-01 20:01:06 +04:00
saharNooby 972e28d48d Implement INT4 conversion and inference 2023-04-01 19:22:01 +04:00
saharNooby b164bf4e27 Allocate memory as needed for specific configuration of model 2023-04-01 17:15:23 +04:00
saharNooby a1e1d34c93 Add Python wrapper for C library 2023-04-01 16:02:22 +04:00
saharNooby 7130a89d1f [FILE FORMAT CHANGED] Reverse dimensions in ggml file (makes it more similar to llama.cpp format) 2023-04-01 14:41:30 +04:00
saharNooby ac03019fcf Move model to separate C library file 2023-04-01 14:38:50 +04:00
saharNooby f6d45baec0 Support FP16 inference 2023-04-01 11:53:49 +04:00
saharNooby fe98c94a63 [FILE FORMAT CHANGED] Use ggml_get_rows to get embedding 2023-04-01 11:28:32 +04:00
saharNooby 16ec7a5c18 Add fail-fast version of the test 2023-04-01 11:15:15 +04:00
saharNooby 0fcb7c64c6 Remove reference implementation code and test against pre-created logits 2023-04-01 11:09:24 +04:00
saharNooby bf88e8a246 Update README.md 2023-04-01 10:12:10 +04:00
saharNooby 6fe9486cee Finally, FP32 inference 2023-04-01 10:06:39 +04:00
saharNooby 61c6b1a4e0 Add comparison against reference implementation script, implement state & logits saving 2023-03-31 20:23:42 +04:00
saharNooby d00f28581a Add reference implementation of RWKV RNN 2023-03-31 19:57:16 +04:00
saharNooby 02c9946b57 Update README.md 2023-03-31 19:06:31 +04:00
saharNooby 01d667f066 Implement exp, max, 1_minus_x, sigmoid operators in ggml 2023-03-31 19:04:35 +04:00
saharNooby fe272dc3d3 Minor changes 2023-03-31 10:24:12 +04:00
saharNooby 93c8dcae75 Update README.md 2023-03-30 20:37:09 +04:00
saharNooby 56bf4fc856 Implement time mixing, fix matrix shape mismatch 2023-03-30 20:29:41 +04:00
saharNooby 873cb954d0 Make ln0 work correctly 2023-03-30 20:01:26 +04:00
saharNooby 2f51451561 Initial commit 2023-03-30 17:55:30 +04:00
slaren ed3c680bcd
Fix GGML_F32Cx8_STORE in AVX without F16C path (#619) 2023-03-30 11:16:30 +02:00
anzz1 9cbc404ba6
ci : re-enable AVX512 testing (Windows-MSVC) (#584) 
* CI: Re-enable AVX512 testing (Windows-MSVC)

Now with 100% less base64 encoding

* plain __cpuid is enough here
 2023-03-29 23:44:39 +03:00
Georgi Gerganov b51c717d5c
ggml : init time on first ggml_init() call 2023-03-29 22:15:34 +03:00
Georgi Gerganov 0ba76c1e73
llama : fix compile warnings when reading the vocab 2023-03-29 22:13:12 +03:00
Georgi Gerganov cea1c85948
ggml : add ARM_NEON dequantize_row_q4_1() 2023-03-29 22:10:01 +03:00
Georgi Gerganov f202ada131
ggml : add ARM_NEON quantize_row_q4_1() 2023-03-29 22:03:07 +03:00
Georgi Gerganov 3b44d30d9b
ggml : add ARM_NEON ggml_vec_dot_q4_1() 2023-03-29 22:03:07 +03:00
Pavol Rusnak 61cbfff5c9
rename convert_ggml_to_pth.py -> convert-ggml-to-pth.py (#600) 
to match filenames of other converters
 2023-03-29 20:09:25 +02:00
Thérence d9ad104440
Create chat-13B.bat (#592) 
* Create chat-13B.bat

Same script than chat-13B.sh, but for windows users.
Tested and working on windows 10/11 v 22H2

* Apply suggestions from code review

---------

Co-authored-by: anzz1 <anzz1@live.com>
 2023-03-29 20:21:09 +03:00
Georgi Gerganov b467702b87
readme : fix typos 2023-03-29 19:38:31 +03:00
Georgi Gerganov 516d88e75c
readme : add GPT4All instructions (close #588) 2023-03-29 19:37:20 +03:00
Georgi Gerganov 53635c081c
py : add GPT4All conversion script 
For now: copy-paste
Too much time for me to deduplicate the python code
 2023-03-29 19:29:52 +03:00
Maël Kerbiriou 41318d708e
llama : use the same threshold for OpenBLAS and ggml thread limiting (#577) 2023-03-29 19:10:07 +03:00
Tobias Lütke a6956b25a1
add example of re-act pattern (#583) 
* add example of re-act pattern

* spelling...

* fixed whitespace in reverse prompt issue
 2023-03-29 10:10:24 -05:00
anzz1 83df5639eb
Fix GCC warning about binary literal (#595) 
0b10101010 -> 0xAA /* 0b10101010 */
 2023-03-29 13:20:07 +00:00
anzz1 a5c42c4b13
Fix typo in llama.h (#593) 2023-03-29 13:19:29 +00:00
anzz1 5a5f8b1501
Enable Fused-Multiply-Add (FMA) and F16C/CVT16 vector extensions on MSVC (#375) 
* Enable Fused-Multiply-Add (FMA) instructions on MSVC

__FMA__ macro does not exist in MSVC

* Enable F16C/CVT16 vector extensions on MSVC

__F16C__ macro does not exist in MSVC, but is implied with AVX2/AVX512

* MSVC cvt intrinsics

* Add __SSE3__ macro for MSVC too because why not

even though it's not currently used for anything when AVX is defined
 2023-03-28 22:44:29 +03:00
anzz1 f1217055ea
CI: fix subdirectory path globbing (#546) 
- Changes in subdirectories will now be detecter properly
- (Windows-MSVC) AVX512 tests temporarily disabled
 2023-03-28 22:43:25 +03:00