tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

bump deepspeed for fix for grad norm compute putting tensors on different devices (#1699)

Browse Source
This commit is contained in:
Wing Lian
2024-06-09 17:13:28 -04:00
committed by GitHub
parent 18cabc0c46
commit 851ccb1237
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
requirements.txt
View File

@@ -5,7 +5,7 @@ transformers==4.41.1
tokenizers==0.19.1 tokenizers==0.19.1
bitsandbytes==0.43.1 bitsandbytes==0.43.1
accelerate==0.30.1 accelerate==0.30.1
deepspeed==0.14.2 deepspeed @ git+https://github.com/microsoft/DeepSpeed.git@bc48371c5e1fb8fd70fc79285e66201dbb65679b
pydantic==2.6.3 pydantic==2.6.3
addict addict
fire fire

2
setup.py
View File

@@ -83,7 +83,7 @@ setup(
"fused-dense-lib @ git+https://github.com/Dao-AILab/flash-attention@v2.5.8#subdirectory=csrc/fused_dense_lib", "fused-dense-lib @ git+https://github.com/Dao-AILab/flash-attention@v2.5.8#subdirectory=csrc/fused_dense_lib",
], ],
"deepspeed": [ "deepspeed": [
"deepspeed==0.14.2", "deepspeed @ git+https://github.com/microsoft/DeepSpeed.git@bc48371c5e1fb8fd70fc79285e66201dbb65679b",
"deepspeed-kernels", "deepspeed-kernels",
], ],
"mamba-ssm": [ "mamba-ssm": [