Commit Graph

166 Commits

Author SHA1 Message Date
Wing Lian
1f5d83ea72 remove un-needed code, add validation 2023-05-24 22:47:43 -04:00
Wing Lian
7e81ca720b Update requirements.txt
Co-authored-by: NanoCode012 <kevinvong@rocketmail.com>
2023-05-24 15:44:48 -04:00
Wing Lian
e8aacfbd7c more qlora support 2023-05-24 14:33:18 -04:00
Wing Lian
b9d07aa95a prepare does all this already for qlora? 2023-05-24 14:32:39 -04:00
Wing Lian
3b4d055edd integrate qlora? maybe? 2023-05-24 14:32:39 -04:00
Wing Lian
2ae936fbc4 fix missing fp16 kwarg 2023-05-23 20:44:24 -04:00
Wing Lian
fb100a9ee1 fix enum pass as value 2023-05-23 11:34:03 -04:00
Wing Lian
3a503770e4 Add qa style data for alpaca instructions, fix one_cycle scheduler 2023-05-22 22:58:10 -04:00
Wing Lian
b029a11e65 Merge pull request #34 from OpenAccess-AI-Collective/dev-unstable
lots of various improvements
2023-05-22 12:14:56 -04:00
Wing Lian
e3df3a9f5d cuda/pytorch matrix builds 2023-05-22 12:14:21 -04:00
Wing Lian
f950a881e1 cuda, pytorch matrix for base builds 2023-05-22 12:12:08 -04:00
Wing Lian
de6da13e19 don't need to set here 2023-05-22 12:12:01 -04:00
Wing Lian
9493b1b137 be able to use adam bnb 8bit and one cycle scheduler w fsdp 2023-05-22 09:00:49 -04:00
Wing Lian
1b3e401241 Update src/axolotl/utils/models.py for info msg
Co-authored-by: NanoCode012 <kevinvong@rocketmail.com>
2023-05-21 23:01:35 -04:00
Wing Lian
3457810988 Update scripts/finetune.py
Co-authored-by: NanoCode012 <kevinvong@rocketmail.com>
2023-05-21 23:00:28 -04:00
Wing Lian
ae1719d30c Update scripts/finetune.py for logging
Co-authored-by: NanoCode012 <kevinvong@rocketmail.com>
2023-05-21 23:00:23 -04:00
Wing Lian
98a6781f18 Update src/axolotl/utils/data.py for spelling
Co-authored-by: NanoCode012 <kevinvong@rocketmail.com>
2023-05-21 23:00:13 -04:00
Wing Lian
607a4d33f2 make sure to use train split if loading from hf 2023-05-21 22:04:39 -04:00
Wing Lian
99383f14a3 make one cycle lr div factor configurable 2023-05-21 20:25:06 -04:00
Wing Lian
0f74464652 fix new dataset prompt tokenizers 2023-05-21 18:57:09 -04:00
Wing Lian
e0602a9e54 add missing __init__ 2023-05-21 16:36:41 -04:00
Wing Lian
2809f3f21b pygmalion dataset prompts format, cached tokenized datasets should be hashed on the tokenizer too 2023-05-21 16:16:09 -04:00
Wing Lian
4ea9a66dbd tokenization fixes 2023-05-21 08:33:06 -04:00
Wing Lian
1d5ab84486 optionally be able to specify alpaca or chat style prompts 2023-05-20 18:16:22 -04:00
Wing Lian
fa8bd14be4 update entrypoint and force min accelerate 2023-05-18 06:25:34 -04:00
Wing Lian
13650732f8 concise multiple choice and tldr summarize 2023-05-17 11:29:17 -04:00
Wing Lian
8c2f3cb0f8 support for replit lm 2023-05-17 08:49:03 -04:00
Wing Lian
b46bc02f0a add alpaca multiple choice instruct dataset support 2023-05-16 21:45:34 -04:00
Wing Lian
f98e173b59 reorder options so debug can happen in the same prepare step 2023-05-15 22:26:30 -04:00
Wing Lian
5e37144754 fix prompters, especially the sharegpt prompter 2023-05-15 22:15:36 -04:00
Wing Lian
bdbca8fa6c more fixes 2023-05-15 14:07:17 -04:00
Wing Lian
42410c783c more fixes 2023-05-14 09:16:41 -04:00
Wing Lian
aef00b6c13 fix torch_dtype for model load 2023-05-14 08:44:22 -04:00
Wing Lian
0d28df0fd2 move filter to before saving so it doesn't happen everytime, update runpod manual script 2023-05-13 21:51:41 -04:00
Wing Lian
84c7bc4b68 whoops, gt vs lt 2023-05-12 14:03:25 -04:00
Wing Lian
aa3c3f97ae optimize dataloading to use cache, fix model token embedding sizes 2023-05-12 13:53:27 -04:00
Wing Lian
f6d1fa4a85 Merge pull request #25 from NanoCode012/patch-2
Fix Trainer() got multiple values for keyword argument 'callbacks'
2023-05-11 09:20:15 -04:00
NanoCode012
89b7f26b9d Merge branch 'main' into patch-2 2023-05-11 21:18:38 +09:00
Wing Lian
165da584b3 fix config for parity with previous change
5159d00a86\#diff-65b4693504c4e8ffac76c7f2c90913faee381f802cf64e7f49c995a2134ed3b3R164
2023-05-11 08:13:09 -04:00
Wing Lian
4cc7ed8898 Merge pull request #27 from NanoCode012/patch-1
Fix save typo
2023-05-11 07:27:31 -04:00
NanoCode012
52aada7174 Fix typo 2023-05-11 20:22:30 +09:00
Wing Lian
688c73a81e Merge pull request #26 from OpenAccess-AI-Collective/mpt-triton
Mpt triton
2023-05-10 16:02:05 -04:00
Wing Lian
2bc1a5bde1 black formatting 2023-05-10 16:01:08 -04:00
Wing Lian
7a490a4646 various fixes 2023-05-10 16:00:09 -04:00
NanoCode012
813aab378f Fix Trainer() got multiple values for keyword argument 'callbacks' 2023-05-10 18:28:28 +09:00
Wing Lian
e2e68c3965 testing mpt triton 2023-05-09 20:57:40 -04:00
Wing Lian
a27d594788 fix conditional so alpaca doesn't choke 2023-05-09 20:57:07 -04:00
Wing Lian
1fb0376150 Merge pull request #23 from NanoCode012/patch-1
Fix: Save adapter for lora
2023-05-09 15:05:58 -04:00
Wing Lian
915c56cd97 Update finetune.py 2023-05-09 15:05:39 -04:00
Wing Lian
df9c5085b5 not everyone has bf16 available 2023-05-09 14:47:48 -04:00