github
/
ChatGLM-6B
mirror of https://github.com/THUDM/ChatGLM-6B
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
332 Commits
5 Branches
0 Tags
26 MiB
Commit Graph

19 Commits (d68ae88488fb247e4f459d50162f4056b97aab08)

Author SHA1 Message Date
lambda d68ae88488 avoid contention in model loading 2023-05-16 07:56:11 +00:00
邓钢清 486edfed81
Update main.py 
fix：修复预测时输入输出长度和超过512时，导致所有预测结果为空的bug
 2023-05-08 10:58:24 +08:00
duzx16 5ade1e4055 Update loading instructions 2023-04-13 16:10:10 +08:00
rainatam 6fb0380847 Merge branch 'main' of https://github.com/THUDM/ChatGLM-6B into main 2023-04-13 13:53:01 +08:00
rainatam 5fb705cd5b Add option for saving checkpoint 2023-04-13 13:52:35 +08:00
duzx16 7607cfe585 Fix turn_idx in eval 2023-04-12 23:41:37 +08:00
duzx16 f06df225dd Fix turn_idx 2023-04-12 23:34:44 +08:00
rainatam cbb9f44e30 Save PrefixEncoder params only 2023-04-10 17:26:17 +08:00
duzx16 ea682a6f51 Update default hyperparameters 
Remove hardcode token id
 2023-04-06 22:42:41 +08:00
rainatam 1cbe2d1981 Remove logging 2023-04-06 20:22:56 +08:00
rainatam 5865924cc6 Add training for chat data 2023-04-06 20:21:29 +08:00
rainatam a9fc018444 Update evaluation results and bleu score function 2023-04-06 15:16:36 +08:00
Qingsong Lv 5de0055408 fix finetune pad bug and add sat readme 2023-04-03 11:27:28 +00:00
duzx16 4227999d4c No padding in colloator 2023-04-02 02:05:03 +08:00
duzx16 ca43864f39 Change quantization instruction 2023-04-02 00:36:14 +08:00
duzx16 4371f7a572 Add padding for evaluation data 2023-04-01 23:09:26 +08:00
maybeluo 7a67ddd61f write generated result with utf-8 2023-04-01 00:34:19 +08:00
duzx16 08d880141d Fix revision for loading model 2023-03-31 16:32:34 +08:00
duzx16 968a30672a Add P-Tuning v2 2023-03-31 10:43:55 +08:00