 Hamid Shojanazeri
|
5b916114eb
merge main branch
|
1 jaar geleden |
|
Hamid Shojanazeri
|
668c364f6b
add rank to save_train_params
|
1 jaar geleden |
|
Hamid Shojanazeri
|
231c9e7da9
adding train_param.yaml saving for fsdp checkpoint loading for inference
|
1 jaar geleden |
|
Hamid Shojanazeri
|
563e572f7c
adding active mem stat
|
1 jaar geleden |
|
Hamid Shojanazeri
|
bd01f64cbd
Merge branch 'main' into fix-cuda_id
|
1 jaar geleden |
 Andrew Gu
|
71fdc4920a
Save memory and fix typos
|
1 jaar geleden |
|
Hamid Shojanazeri
|
a7156dfb5d
fixing the cuda id
|
1 jaar geleden |
|
Hamid Shojanazeri
|
707af7ea24
adding cuda:0 for non-fsdp situations
|
1 jaar geleden |
|
Hamid Shojanazeri
|
6678be75ad
fixing identation
|
1 jaar geleden |
|
Hamid Shojanazeri
|
6a84e9e4d5
fixing scaler for both fsdp and non fsdp
|
1 jaar geleden |
|
Hamid Shojanazeri
|
065ddaa77b
fixing the condition for moving to cuda
|
1 jaar geleden |
|
Hamid Shojanazeri
|
20b061e01c
modify to steping the lr scheduler each epoch
|
1 jaar geleden |
 chauhang
|
4767f09ecd
Initial commit
|
1 jaar geleden |
