| .. |
|
tools
|
integrated world tokenizer for RWKV
|
2023-06-13 20:06:19 +08:00 |
|
ggml_v1.c
|
just testing cublas
|
2023-05-15 20:01:22 +08:00 |
|
ggml_v1.h
|
Various enhancement and integration pygmalion.cpp
|
2023-04-03 00:04:43 +08:00 |
|
ggml_v2-cuda-legacy.cu
|
fix compile errors
|
2023-06-29 17:54:12 +08:00 |
|
ggml_v2-cuda-legacy.h
|
Wip, CUDA porting malloc improvements, gpu accel for non-llama, backport old quants
|
2023-06-28 18:20:46 +08:00 |
|
ggml_v2-cuda.cu
|
cleanup indentation, fixing cublas build
|
2023-06-08 22:40:53 +08:00 |
|
ggml_v2-cuda.h
|
cleanup indentation, fixing cublas build
|
2023-06-08 22:40:53 +08:00 |
|
ggml_v2-opencl-legacy.c
|
Merge remote-tracking branch 'occam/opencl-dev' into concedo_experimental
|
2023-05-22 16:16:48 +08:00 |
|
ggml_v2-opencl-legacy.h
|
still merging in process
|
2023-05-20 15:58:33 +08:00 |
|
ggml_v2-opencl.cpp
|
back to 32 block size, not better
|
2023-06-01 00:14:22 +08:00 |
|
ggml_v2-opencl.h
|
still merging in process
|
2023-05-20 15:58:33 +08:00 |
|
ggml_v2.c
|
Wip, CUDA porting malloc improvements, gpu accel for non-llama, backport old quants
|
2023-06-28 18:20:46 +08:00 |
|
ggml_v2.h
|
remove old filever checks (+7 squashed commit)
|
2023-05-21 00:15:39 +08:00 |
|
gpt2_v1.cpp
|
fixed other arch
|
2023-05-24 00:20:43 +08:00 |
|
gpt2_v2.cpp
|
cleaning up some old junk
|
2023-06-04 11:05:46 +08:00 |
|
gpt2_v3.cpp
|
fix for yr-rocm, large gpu scratch
|
2023-06-30 12:40:08 +08:00 |
|
gptj_v1.cpp
|
remove old filever checks (+7 squashed commit)
|
2023-05-21 00:15:39 +08:00 |
|
gptj_v2.cpp
|
cleaning up some old junk
|
2023-06-04 11:05:46 +08:00 |
|
gptj_v3.cpp
|
fix for yr-rocm, large gpu scratch
|
2023-06-30 12:40:08 +08:00 |
|
llama_v2-util.h
|
cleanup indentation, fixing cublas build
|
2023-06-08 22:40:53 +08:00 |
|
llama_v2.cpp
|
fix for yr-rocm, large gpu scratch
|
2023-06-30 12:40:08 +08:00 |
|
llama_v2.h
|
still merging in process
|
2023-05-20 15:58:33 +08:00 |
|
mpt_v3.cpp
|
fix for yr-rocm, large gpu scratch
|
2023-06-30 12:40:08 +08:00 |
|
neox_v2.cpp
|
fixed other arch
|
2023-05-24 00:20:43 +08:00 |
|
neox_v3.cpp
|
fix for yr-rocm, large gpu scratch
|
2023-06-30 12:40:08 +08:00 |
|
otherarch.h
|
Implemented basic GPU offloading for MPT, GPT-2, GPT-J and GPT-NeoX
|
2023-06-22 00:43:25 +08:00 |
|
rwkv_v2.cpp
|
wip integrating new rwkv
|
2023-05-27 22:45:28 +08:00 |
|
rwkv_v2.h
|
wip integrating new rwkv
|
2023-05-27 22:45:28 +08:00 |
|
rwkv_v3.cpp
|
reverted sequence mode for rwkv due to multiple issues with speed loss with bigger quantized models
|
2023-06-14 20:03:14 +08:00 |
|
rwkv_v3.h
|
rwkv compile fix (+1 squashed commits)
|
2023-06-12 23:04:40 +08:00 |
|
rwkv_vocab.cpp
|
integrated world tokenizer for RWKV
|
2023-06-13 20:06:19 +08:00 |
|
utils.cpp
|
wstring convert for mpt
|
2023-06-24 11:43:42 +08:00 |
|
utils.h
|
wstring convert for mpt
|
2023-06-24 11:43:42 +08:00 |
