Skip to content

fix: seed RNG in nemotron_h flaky test #2157

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鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mikasenghaas merged 1 commit into from
Mar 31, 2026
Merged

fix: seed RNG in nemotron_h flaky test #2157

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
1 change: 1 addition & 0 deletions tests/unit/train/models/test_nemotron_h.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,6 +84,7 @@ def test_nemotron_h_mamba_moe_only():
if isinstance(layer, NemotronHAttentionLayer):
layer.forward = lambda hidden_states, **kwargs: hidden_states

torch.manual_seed(42)

@cursor cursor Bot Mar 31, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Global RNG state leaks across tests

Medium Severity

torch.manual_seed(42) mutates the process-wide RNG state and this test does not restore it. That makes later tests inherit a changed random stream, creating order-dependent behavior and potential flakiness outside test_nemotron_h_mamba_moe_only.

Fix in Cursor聽Fix in Web

with torch.device("cuda"), default_dtype(torch.float32):
input_ids = torch.randint(0, 256, (1, 32))
position_ids = torch.arange(0, 32).unsqueeze(0)
Expand Down
Loading