You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,6 +1,6 @@
1
1
# OpenDiLoCo
2
2
3
-
This repository contains the training code and experiment results for the paper "OpenDiLoCo: An Open-Source Framework for Globally Distributed Low-Communication Training".
3
+
This repository contains the training code and experiment results for the paper [OpenDiLoCo: An Open-Source Framework for Globally Distributed Low-Communication Training](https://arxiv.org/abs/2407.07852).
4
4
5
5
# Setup
6
6
@@ -303,4 +303,4 @@ We recommend using `bf16` to avoid scaling and desynchronization issues with hiv
303
303
304
304
3. `torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate...`
305
305
A possible culprit is that your `--per-device-train-batch-size` is too high.
0 commit comments