josc146
|
f05a4acb04
|
sync https://github.com/JL-er/RWKV-PEFT
|
2024-05-28 22:35:47 +08:00 |
|
josc146
|
0e4b6cbd15
|
make gate and out trainable (834aea0f54)
|
2024-03-24 15:47:17 +08:00 |
|
josc146
|
c5077f4ebc
|
fix v6 lora (c03cdbbdaf)
|
2024-03-14 12:25:09 +08:00 |
|
josc146
|
5692579f56
|
for Chinese users, replace Tsinghua pip mirrors with Alibaba Cloud to avoid 403 http error
|
2024-03-13 21:37:35 +08:00 |
|
josc146
|
333619839a
|
rwkv6 lora finetune support (https://github.com/JL-er/RWKV-LORA)
|
2024-03-13 17:51:53 +08:00 |
|
josc146
|
e0a6a279b3
|
add python3-dev to lora fine-tune dependencies
|
2024-02-28 23:34:49 +08:00 |
|
josc146
|
0da92ec7bf
|
improve fine-tune performance
|
2024-02-04 19:33:32 +08:00 |
|
josc146
|
81544ca8b3
|
rwkv5 lora finetune support (https://github.com/JL-er/RWKV-v5-lora)
|
2023-12-29 12:23:36 +08:00 |
|
josc146
|
a8b4f0bb7e
|
lora finetune version check
|
2023-11-30 13:01:38 +08:00 |
|
josc146
|
f739c61197
|
fix a finetune bug
|
2023-11-17 22:37:21 +08:00 |
|
josc146
|
d249a4c29a
|
print error.txt
|
2023-11-08 22:57:38 +08:00 |
|
josc146
|
b5a6f8a425
|
set deepspeed to 0.11.2 to avoid finetune error
|
2023-11-08 22:20:11 +08:00 |
|
josc146
|
1ad86d737c
|
chore
|
2023-11-08 22:18:49 +08:00 |
|
josc146
|
1d7f19ffaf
|
update sample.jsonl
|
2023-10-26 14:08:16 +08:00 |
|
josc146
|
fe0860dbf0
|
fix lora finetune max_epochs (#170)
|
2023-08-24 22:49:57 +08:00 |
|
josc146
|
02d5d641d1
|
chore
|
2023-08-24 22:48:54 +08:00 |
|
josc146
|
5ee5fa7e6e
|
fix load_state_dict crash
|
2023-07-09 12:33:29 +08:00 |
|
josc146
|
d8c70453ec
|
format
|
2023-07-09 12:32:50 +08:00 |
|
josc146
|
6fbb86667c
|
improve python script error messages
|
2023-07-07 20:16:35 +08:00 |
|
josc146
|
55210c89e2
|
improve wsl dependencies installation
|
2023-07-07 18:57:51 +08:00 |
|
josc146
|
511652b71c
|
improve finetune compatibility
|
2023-07-03 22:19:20 +08:00 |
|
josc146
|
76761ee453
|
improve lora finetune process (need to be refactored)
|
2023-07-03 21:40:16 +08:00 |
|
josc146
|
987854fe49
|
lora finetune (need to be refactored)
|
2023-07-03 17:41:47 +08:00 |
|