Commit Graph

37 Commits

Author SHA1 Message Date
Wing Lian
a12fb0a8da Jeopardy bot! (#17)
* support for jeopardy dataset

* commit the final config for jeopardy bot
2023-05-08 03:21:40 -04:00
Wing Lian
a4329b1068 fix #16 load best model setting when using 8bit 2023-05-07 18:30:48 -04:00
Wing Lian
550502b321 use micro batch size for eval size if not specified 2023-05-07 18:26:05 -04:00
Wing Lian
247825bd57 refactor inference, warn if model is frozen 2023-05-07 01:54:15 -04:00
Wing Lian
cb9a887047 Merge pull request #13 from winglian/dev
merge dev branch for various fixes
2023-05-07 01:48:02 -04:00
NanoCode012
0e74b6402e Add eval_batch_size for evaluation 2023-05-06 22:21:24 +09:00
Wing Lian
a10a8265ef fix log sweep lr 2023-05-03 15:06:03 -04:00
Wing Lian
9105935b00 support for multi line inference input, log sweep over learning rates 2023-05-03 13:48:54 -04:00
Wing Lian
7748f3d6da fix adam bnb optimizer grouped parameters, fix peft model 8bit conversion logic, black formatting 2023-05-01 16:31:46 -04:00
Wing Lian
2255bb7f4f support llama-adapter zero init attention 2023-05-01 10:42:21 -04:00
Wing Lian
ad2b48c0fa fdsp config dict fix, todo list, add torchdistx support 2023-04-30 13:32:07 -04:00
Wing Lian
9190ada23a 8bit and deepspeed changes 2023-04-30 06:50:35 -04:00
Wing Lian
6dfdd2dec0 don't load models in 8bit unless they are using an adapter, also fix tokenizer load in exceptional case 2023-04-30 03:19:56 -04:00
Wing Lian
29936bba7f fix fsdp training args 2023-04-30 00:56:28 -04:00
Wing Lian
78821815de fix for zero value warmup steps 2023-04-30 00:34:12 -04:00
Wing Lian
5159d00a86 fix sharegpt tokenization, refactor tokenization debugging 2023-04-30 00:23:53 -04:00
Wing Lian
c0f50d9c61 wire up gradient checkpointing for 4bit 2023-04-28 22:28:41 -04:00
Wing Lian
4a17a4c9a1 fix dataset handling, support galactica 2023-04-24 10:54:45 -04:00
Wing Lian
097d367af6 tweaks to data loading, 8 bit adam, accelerate and deepspeed 2023-04-24 09:41:35 -04:00
Wing Lian
4f2584f2dc shuffle and split dataset after save/load 2023-04-24 09:41:35 -04:00
Wing Lian
8d437853c8 fix sharegpt handling from hf, don't worry about loading llama if using earlier transformers release 2023-04-24 09:41:35 -04:00
Wing Lian
94f5e415a3 various bugfixes 2023-04-24 09:41:34 -04:00
Wing Lian
bb991fd870 fix bug when model_type not explicitly passed 2023-04-19 13:15:33 -04:00
Wing Lian
d65385912e improve inference 2023-04-19 12:57:27 -04:00
Wing Lian
0a472e1e08 quickstart instructions for starting from runpod (#5) 2023-04-18 19:22:25 -04:00
Wing Lian
8746b701fe attempt xformers hijack attention 2023-04-18 14:03:50 -04:00
Wing Lian
6045345d6b WIP large refactor to make finetune script a little more manageable (#3) 2023-04-18 14:01:38 -04:00
Wing Lian
81de0efc18 add support for alpaca reflect training (#2) 2023-04-18 08:34:05 -04:00
Wing Lian
87d7825435 Tokenization open assistant (#1)
* refactor prompt tokenization to more easily support open assistant

* add open assisstant handling, more logging, black formatting
2023-04-18 01:45:49 -04:00
Wing Lian
e1076430ff suppport for alpaca-like instruction datasets without inputs 2023-04-17 23:32:57 -04:00
Wing Lian
2db9436410 casts the prepared data to int16 (doesn't help with training memory) 2023-04-17 21:36:02 -04:00
Wing Lian
77fca25f1b 4bit quantized support (wip) 2023-04-17 11:37:39 -04:00
Wing Lian
80b2ed29d8 various bugfixes 2023-04-14 21:37:07 -04:00
Wing Lian
f2a2029d0d config chooser, update readme instructions, device config, llama flash attention, debug out the labels, fix config key checks, other bugfixes 2023-04-14 12:18:56 -04:00
Wing Lian
a6028d302e black formatting 2023-04-14 07:25:52 -04:00
Wing Lian
8d959a7e26 make it work with pythia in the cloud 2023-04-14 07:24:55 -04:00
Wing Lian
ce24f5e246 WIP for axolotl trainer 2023-04-14 00:20:05 -04:00