Skip to content

Fix: Handle div_scale when using Gemini plugin with CPUAdam #6415

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Truong5724 wants to merge 2 commits into
base: main
Choose a base branch
from
+6 −2
Open

Fix: Handle div_scale when using Gemini plugin with CPUAdam #6415

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
969885e
Truong Handle div_scale in CPUAdam Optimizer
Truong5724 Apr 2, 2026
f0a699f
Merge branch 'hpcaitech:main' into Truong-Fix-CPUAdam
Truong5724 Apr 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions colossalai/nn/optimizer/cpu_adam.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -190,17 +190,21 @@ def step(self, closure=None, div_scale: float = -1):
)
self._post_update(p, "exp_avg", "exp_avg_sq")
elif target_device.type == "cuda":
assert div_scale == -1, "div_scale should remain default"
assert state["exp_avg"].device.type == "cuda", "exp_avg should stay on cuda"
assert state["exp_avg_sq"].device.type == "cuda", "exp_avg should stay on cuda"

bias_correction1 = 1 - beta1 ** state["step"]
bias_correction2 = 1 - beta2 ** state["step"]

# scale gradient if div_scale is provided
grad = p.grad.data
if div_scale != -1:
grad = grad / div_scale

# adam on cuda
self.torch_adam_update(
p.data,
p.grad.data,
grad,
state["exp_avg"],
state["exp_avg_sq"],
group["lr"],
Expand Down