90fe9096b4
clean and refactoring pass before supporting newer models for different arch
2023-05-17 11:23:29 +08:00
94ef3e81cf
inc allocation
2023-05-16 23:32:35 +08:00
72836d4eac
fixing more compile issues
2023-05-15 20:10:54 +08:00
6504150fac
just testing cublas
2023-05-15 20:01:22 +08:00
c81dd58e76
Merge commit 'f954edda935a70a14cf0cc45ecc7fe7d60cf3e4b' into archive_lib
...
# Conflicts:
# ggml.c
2023-05-14 18:34:56 +08:00
b692e4d2a4
wip
2023-05-14 17:21:07 +08:00
e05455f852
fixed wrong sized struct from legacy q8_1, fixed opencl varsize arrays
2023-05-13 23:56:08 +08:00
05cf5f7d6e
partially working, but the blas matmul is broken
2023-05-13 11:35:38 +08:00
b335f73a60
BACKWARDS COMPAT QUANT SHIM is ready, but upstream model converter is BORKED. BORK BORK.
2023-05-13 01:30:11 +08:00
e47f7ade05
updated kobold lite, patch oom errors
2023-05-09 19:16:45 +08:00
2f2eff6e13
the dark gods have been sated, and redpajama is integrated... but at what cost?
2023-05-08 20:58:00 +08:00
105f818d45
integrated new version of rwkv from upstream
2023-05-03 23:26:39 +08:00
3de34ee492
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# CMakeLists.txt
# Makefile
# ggml-opencl.c
2023-05-01 12:03:46 +08:00
b3315459c7
pilled the new dequants for clblast, fixed some ooms
2023-04-30 14:15:44 +08:00
032a171867
integrated q5 formats
2023-04-28 12:58:39 +08:00
5070815dcf
fixing discussion #121 and issue #122
2023-04-27 16:10:01 +08:00
0aa3d839fb
free old ctx on retry
2023-04-25 23:42:57 +08:00
72b2331ad6
edge cases with mem crash? need verify
2023-04-25 20:42:30 +08:00
5eec5d6ed9
Added backwards compatibility to an earlier version of NeoX.
2023-04-25 20:34:18 +08:00
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
432cc91649
still needs to be a bit higher for very small contexts
2023-04-23 15:01:38 +08:00
4e1ea2ac61
hopefully fixed the ooms for good
2023-04-23 13:49:50 +08:00
d41490c27b
just revert back to the working commit
2023-04-23 00:35:42 +08:00
c60fb5ef4b
fixed rwkv build errors on ARM devices
2023-04-23 00:18:38 +08:00
b5d6284190
increase initial buffer too
2023-04-23 00:07:33 +08:00
d2f14b2b1f
add an extra buffer to mem allocations
2023-04-23 00:04:32 +08:00
c454f8b848
Gpt NeoX / Pythia integration completed
2023-04-22 11:23:25 +08:00
4fa3dfe8bc
just doesn't work properly on windows. will leave it as a manual flag for others
2023-04-22 10:57:38 +08:00
68898046c2
accidentally added the binaries onto repo again.
2023-04-22 00:41:19 +08:00
7ba36c2c6c
trying to put out penguin based fires. sorry for inconvenience
2023-04-20 23:15:07 +08:00
49697d86d8
adjusted down the buf memory allocation now that realloc seems to work
2023-04-20 17:51:13 +08:00
cc407f283a
messing around with memory allocation to bandaid the random ooms with various gpt2 and gptj models
2023-04-19 20:18:55 +08:00
45ec09d31b
fast forwarding for rwkv for unmodified contexts
2023-04-19 15:09:35 +08:00
ea01771dd5
rwkv is done
2023-04-18 20:55:01 +08:00
c200b674f4
updated kobold lite, work on rwkv, added exe path to model load params, added launch parameter
2023-04-18 17:36:44 +08:00
763ad172c0
arranged files, updated kobold lite, modified makefile for extra link args on linux, started RWKV implementation
2023-04-17 17:31:45 +08:00
c757fbee1d
fixes to stopper tokens, fixed BLAS mode for GPT2 and GPTJ, updated kobold lite
2023-04-16 21:54:18 +08:00
1bd5992da4
clean and refactor handling of flags
2023-04-12 23:25:31 +08:00
636f8e5a8e
updated the quantize files and makefile
2023-04-12 21:40:25 +08:00
69b85f5b61
fixed a few OOM errors with larger contexts - I cannot figure out why they happen, so I am forced to increase the buffer size.
2023-04-11 00:14:57 +08:00
18a154715e
added version label, improved file type checks
2023-04-10 01:03:09 +08:00
d8e37bfe75
new gpt2 format supported
2023-04-08 17:35:36 +08:00
1abcdb2394
should not be static
2023-04-07 20:35:19 +08:00
1d48db4f63
dont build quantize
2023-04-07 17:11:26 +08:00
4f5faf9612
some users report that this repo is now being flagged as malicious?
...
no idea why, but I am removing all prebuilt binaries except libopenblas. windows users can still obtain it from /releases and osx and linux users can rebuild from source code.
2023-04-06 21:49:43 +08:00
3d650d0e25
remove dependency of psutil, fixed compile error on WSL, handle exceptions when sending http response, added multiline for embedded kobold
2023-04-06 11:08:19 +08:00
1490cdd71d
change GPT-J and GPT2 KVs to use fp16 instead
2023-04-05 15:53:07 +08:00
57e9f929ee
renamed misnamed ACCELERATE define, and removed all -march=native and -mtune=native flags
2023-04-05 15:22:13 +08:00
14273fea7a
integrated gpt2 support
2023-04-04 23:15:47 +08:00
52de932842
removed main.exe to reduce clutter, added support for rep pen in gptj
2023-04-04 20:43:13 +08:00
Concedo