Wrong Weight Initialization
Wrong weight initialization causes training instability with NaN losses, slow convergence, or dead neurons.
Wrong weight initialization causes training instability with NaN losses, slow convergence, or dead neurons.
What this failure is
Wrong Weight Initialization is a Training Stability failure seen during ML training runs. Wrong weight initialization causes training instability with NaN losses, slow convergence, or dead neurons. Common tags: Weight Initialization, Kaiming, Xavier, Training Stability.
Is this what broke your run? Paste your log.
You're reading about Wrong Weight Initialization. Paste your own crash log or traceback below and get the real root cause for YOUR run, not this generic entry. No account, no card. Logs are masked at ingress and never saved to account history.
Want 14 days on the Scale plan?
Request an evaluation code. A verified workplace organization activates up to 50 diagnoses a day, alerts, history, and follow-up questions. No credit card or automatic subscription.
Why it happens (the mechanism)
Weights initialized too small causing vanishing gradients. Weights initialized too large causing exploding gradients. Activation function not suited to initialization scheme. No weight initialization specified in model code. Taken together, these mechanisms explain why the failure is reproducible, why it tends to surface on specific workloads or scales, and why generic mitigation attempts often fall short without addressing the underlying cause.
What you'll observe
- Loss is NaN from the first step
- Loss decreases very slowly or not at all
- Many neurons output 0 (dead ReLU)
Common symptoms and what they mean
| Symptom | Why it happens |
|---|---|
| loss: nan in first step | Weights initialized too small causing vanishing gradients |
| loss stays constant for thousands of steps | Weights initialized too large causing exploding gradients |
| Model produces constant outputs for all inputs | Activation function not suited to initialization scheme |
Which systems are affected
- Training from scratch (not fine-tuning)
- Training with new model architecture
- Models with deep networks
How to confirm this is the problem
Use this checklist to test the hypothesis against a small reproduction. No single line proves the root cause, so preserve the preceding events and compare one variable at a time.
- ✓Reproduce the failure from a clean checkpoint/seed: the symptom must appear without warm-up state from a previous run.
- ✓Verified signal present: loss: nan in first step
- ✓Verified signal present: loss stays constant for thousands of steps
- ✓Verified signal present: Model produces constant outputs for all inputs
- ✓A targeted fix from the "How to fix it" section eliminates or substantially reduces the symptom within one validation pass.
The fix and the prevention pattern
The root cause is on this page and stays free. A free account adds the exact remediation steps, saved history, and the fix on every entry in the encyclopedia.
Sign up free. Unlock the full analysisNo credit card. Daily allowance follows verified trust tier. Instant access.
Diagnose this failure in VS Code
Select the traceback or open the failed terminal, then run Denpex locally to see the initiating rank, collateral failures, exact fix, and verification command without uploading the log.
Install the free VS Code extensionRelated failures to investigate next
Root cause
- Weights initialized too small causing vanishing gradients
- Weights initialized too large causing exploding gradients
- Activation function not suited to initialization scheme
- No weight initialization specified in model code
The fix and how to prevent it
Evaluate Denpex on your own logs
Request a Scale evaluation code. A verified workplace organization activates 14 days with up to 50 diagnoses a day. Every account keeps its current diagnosis allowance and gets a verification path. No card or automatic subscription.
Don't just read the fix, diagnose your run
The encyclopedia tells you what went wrong. Denpex tells you what went wrong in YOUR training run. With your logs, your config, and your stack.