Commit Graph

36 Commits

Author SHA1 Message Date
Zhengxiao Du 02a065cf27 Upload pytorch_model.bin 2023-05-15 12:41:28 +00:00
Zhengxiao Du e214c5b71d Update slack link 2023-05-12 13:49:56 +00:00
duzx16 d8a6cfc6cb Update decode method in tokenizer 2023-05-09 11:32:40 +08:00
duzx16 f6b88da8c1 Add support for parallel quantization on Mac 2023-05-04 21:45:31 +02:00
duzx16 63d66b0572 Remove assert in load_cpu_kernel 2023-04-29 10:34:45 +08:00
duzx16 f55a1089a2 Sync with chatglm-6b 2023-04-28 20:17:25 +08:00
duzx16 e02ba894cf Remove pytorch_model.bin.index.json 2023-04-17 21:34:20 +08:00
duzx16 6498797e79 Update slack link 2023-04-17 16:08:53 +08:00
duzx16 1e40d965fe Add pytorch_model.bin.index.json 2023-04-16 21:04:30 +08:00
songxxzp 630d0efd8b Add assertion when loading cpu and cuda kernel fails 2023-04-14 20:00:41 +08:00
songxxzp bcc35f08b4 Add assertion when loading cpu and cuda kernel fails 2023-04-14 19:58:42 +08:00
songxxzp fe0674f86d Merge branch 'dev' 2023-04-14 19:01:28 +08:00
songxxzp c7d8998bb3 Update CPU kernel loading method 2023-04-14 18:52:35 +08:00
duzx16 3485994337 Fix gmask 2023-04-14 15:57:11 +08:00
duzx16 9333486c30 Add empty_init option 2023-04-13 20:35:45 +08:00
duzx16 6466cdcff5 Update README.md 2023-04-13 15:44:48 +08:00
duzx16 9163f7e6d9 Fix eos token in tokenizer 2023-04-11 13:25:12 +08:00
duzx16 649466f2d7 Update dependency 2023-04-09 20:34:34 +08:00
duzx16 41fda88421 Fix attention score on mps 2023-04-09 20:30:23 +08:00
duzx16 a7272d4c93 Fix logit processor
Fix tokenizer config saving
2023-04-08 12:07:06 +08:00
duzx16 96de7a2fb3 Merge branch 'slim' of https://huggingface.co/THUDM/chatglm-6b-int4 into slim 2023-04-07 23:36:57 +08:00
duzx16 5fc46d22f7 Fix embedding quantization 2023-04-07 23:34:41 +08:00
Zhengxiao Du 7edbdfecad Upload pytorch_model.bin 2023-04-07 15:20:38 +00:00
duzx16 bfb1a8f2b6 Slim embedding 2023-04-07 23:14:10 +08:00
songxxzp 68873da921 Fix bugs when compiling cpu kernels 2023-04-06 20:09:56 +08:00
duzx16 1f34060390 Drop icetk dependency 2023-04-06 19:24:30 +08:00
duzx16 19685a5a7e Fix position ids expand 2023-04-03 14:16:48 +08:00
songxxzp 7aaf3fe491 Synchronize with chatglm 6b repo 2023-04-03 09:20:14 +08:00
songxxzp 7458231b5a Fix parallel cpu kernel 2023-04-01 21:15:30 +08:00
songxxzp dac03c3ac8 Fix bugs in quantization when loading kernels 2023-03-22 05:18:51 -04:00
duzx16 debaf0032c Fix Chinese punctuation 2023-03-22 14:49:29 +08:00
Zeng Aohan 3ba9437241 Update README.md 2023-03-20 06:51:58 +00:00
Zeng Aohan 0d0e8067f8 Update README.md 2023-03-20 06:43:49 +00:00
Zeng Aohan 7ad727c89f Update README.md 2023-03-20 06:43:29 +00:00
Sengxian a93efa90f5 init commmit 2023-03-19 21:15:20 +08:00
Zhengxiao Du 62a9758691 initial commit 2023-03-19 12:01:56 +00:00