163 commits

Author	SHA1	Message	Date
Concedo	bfeb3471d7	fix typos	2023-07-03 21:36:42 +08:00
Concedo	3d2907d208	make gptneox and gptj work with extended context too	2023-07-02 18:28:09 +08:00
Concedo	ef3b8dc0d9	GPU accel for rwkv is slow, disable it	2023-07-02 00:41:46 +08:00
Concedo	e1a7042943	try out the new rwkv but it seems worse, may revert	2023-07-02 00:10:56 +08:00
Concedo	86469d15c4	fix for yr-rocm, large gpu scratch	2023-06-30 12:40:08 +08:00
Concedo	86b061b98c	wip on unified cublas integration, add all the small libraries but exclude the large ones	2023-06-29 18:35:31 +08:00
Concedo	c2f1ed6556	fix compile errors	2023-06-29 17:54:12 +08:00
Concedo	b4698abafc	Wip, CUDA porting malloc improvements, gpu accel for non-llama, backport old quants	2023-06-28 18:20:46 +08:00
Concedo	9527a783ea	fix rope inplace	2023-06-27 19:44:33 +08:00
Concedo	8342fe81b1	revert the wstring tokenization. coherency was affected	2023-06-24 12:58:49 +08:00
Concedo	0485fa65a2	wstring convert for mpt	2023-06-24 11:43:42 +08:00
Concedo	490cf395f8	better alloc error	2023-06-23 22:51:51 +08:00
Concedo	f39a746089	bug fixes for openblas	2023-06-23 22:45:22 +08:00
Concedo	43c2891afa	option to not use scratch	2023-06-23 19:01:36 +08:00
Concedo	d5e4cf7ffe	handle ctx manip	2023-06-23 19:01:15 +08:00
Concedo	df9135e3a9	fixing memory bugs	2023-06-23 18:41:23 +08:00
Concedo	e6ddb15c3a	cleanup	2023-06-22 10:38:27 +08:00
Concedo	1b71752a9f	Implemented basic GPU offloading for MPT, GPT-2, GPT-J and GPT-NeoX	2023-06-22 00:43:25 +08:00
Concedo	dfdd20240c	gpt j use scratch buffers	2023-06-21 16:10:31 +08:00
Concedo	8e2dc19dc6	updated tokenizer, added support for scratch buffers for neox and gpt2	2023-06-19 21:29:06 +08:00
Concedo	3ed3e7b7e2	reverted sequence mode for rwkv due to multiple issues with speed loss with bigger quantized models	2023-06-14 20:03:14 +08:00
Concedo	871009dfab	integrated world tokenizer for RWKV	2023-06-13 20:06:19 +08:00
Concedo	860fb026df	rwkv compile fix (+1 squashed commits) ... Squashed commits: [8b0ebb1] upgraded rwkv + added memory overheads + added state_out bufs	2023-06-12 23:04:40 +08:00
Concedo	c44b9c3ecf	added the llama_v2 cuda back (+2 squashed commit) ... Squashed commit: [1c97fd4] Revert "fix for cublas" This reverts commit 994be9a4db. [fce03c3] Revert "fix for cublas" This reverts commit 33528f5b1d.	2023-06-11 23:23:24 +08:00
Concedo	a6a0fa338a	cleanup indentation, fixing cublas build	2023-06-08 22:40:53 +08:00
Concedo	c046db5197	lite bugfixes, buffer size changes, fixed a topk bug.	2023-06-06 22:38:25 +08:00
Concedo	9270056269	fixed compile error in cmake VS	2023-06-05 11:48:04 +08:00
Concedo	9aa2d8535b	hide gpu input box when dropdown not selected, minor memory fix for neox and gptj	2023-06-04 21:47:17 +08:00
Concedo	20803c221e	cleaning up some old junk	2023-06-04 11:05:46 +08:00
Concedo	b62279cb39	buf size for starcoder still not good	2023-06-04 00:41:08 +08:00
Concedo	c1b293d31a	fixed MPT ooms	2023-06-03 18:37:13 +08:00
Concedo	6f82e17b7a	added MPT support	2023-06-03 16:14:08 +08:00
Concedo	234270bd83	back to 32 block size, not better	2023-06-01 00:14:22 +08:00
Concedo	446e42a8c6	change dmmv block size	2023-05-31 21:40:12 +08:00
Concedo	6b3373cb81	revert bad fix	2023-05-29 22:06:12 +08:00
Concedo	ef16d09a51	fix for older gcc, updated lite	2023-05-29 18:54:15 +08:00
Concedo	97b39f875c	fixed fstat64 build error on mac	2023-05-29 15:50:07 +08:00
Concedo	55e0fbf024	wip integrating new rwkv	2023-05-27 22:45:28 +08:00
Concedo	6d7749c98f	no difference	2023-05-27 12:42:19 +08:00
Concedo	bd4fe936f5	cleanup sampling code	2023-05-27 11:58:39 +08:00
Concedo	bf482d1786	revert klite newline bug, trying to add win7 support	2023-05-24 22:21:01 +08:00
Concedo	844f92688a	subpattern fix	2023-05-24 16:48:39 +08:00
Concedo	abb9ad789c	fixed other arch	2023-05-24 00:20:43 +08:00
Concedo	981d5ba866	Merge remote-tracking branch 'occam/opencl-dev' into concedo_experimental ... # Conflicts: # .github/workflows/build.yml # CMakeLists.txt # Makefile # README.md # ggml-opencl.cpp # llama.cpp # otherarch/ggml_v2-opencl-legacy.c	2023-05-22 16:16:48 +08:00
Concedo	587308a202	fixed some build errors on linux, changed icon resolution, added more error printing	2023-05-22 12:18:42 +08:00
Concedo	fea84c3cf5	fix for stupid msvc compiler	2023-05-21 22:41:33 +08:00
Concedo	60e0c67874	fix compile errors on cuda	2023-05-21 21:13:17 +08:00
Concedo	33528f5b1d	fix for cublas	2023-05-21 21:03:36 +08:00
Concedo	994be9a4db	fix for cublas	2023-05-21 21:02:21 +08:00
Concedo	24127ebf98	updated lite, fixed some encoding issues	2023-05-21 17:29:00 +08:00