File: leanpass/nn.py
Dropout.forward scales the mask by 1/(1-p). When p is very close to 1 (e.g., 0.9999) the scaling factor becomes huge, leading to overflow or exploding activations.
Fix: enforce a stricter upper bound (e.g., p <= 0.9) or use a numerically stable inverted‑dropout implementation that caps the scaling factor.
Label: bug
Filed automatically by ai-issue-scan.
File:
leanpass/nn.pyDropout.forwardscales the mask by1/(1-p). Whenpis very close to 1 (e.g., 0.9999) the scaling factor becomes huge, leading to overflow or exploding activations.Fix: enforce a stricter upper bound (e.g.,
p <= 0.9) or use a numerically stable inverted‑dropout implementation that caps the scaling factor.Label: bug
Filed automatically by ai-issue-scan.