Silent Data Corruption: Silent Degradation (No NaN)
A faulty GPU corrupts computation without ever producing a NaN. Loss settles slightly above baseline and parameters drift, making it the most deceptive SDC mode.
A faulty GPU corrupts computation without ever producing a NaN. Loss settles slightly above baseline and parameters drift, making it the most deceptive SDC mode.
What this failure is
Silent Data Corruption: Silent Degradation (No NaN) is a Data Integrity failure seen during ML training runs. A faulty GPU corrupts computation without ever producing a NaN. Loss settles slightly above baseline and parameters drift, making it the most deceptive SDC mode. Common tags: Sdc, Silent Corruption, Data Integrity, Parameter Drift.
Is this what broke your run? Paste your log.
You're reading about Silent Data Corruption: Silent Degradation (No NaN). 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)
Hardware fault flips bits in computation without triggering NaN/Inf. Errors accumulate as parameter drift, not crashes. NaN checks don't catch sub-NaN corruption. 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
- Model trains 'fine' but underperforms
- No NaN, no crash, no error
- Final quality is quietly worse than expected
Common symptoms and what they mean
| Symptom | Why it happens |
|---|---|
| Loss settles above baseline with no NaN events | Hardware fault flips bits in computation without triggering NaN/Inf |
| Parameters drift from a reference run | Errors accumulate as parameter drift, not crashes |
| Validation metrics quietly degrade | NaN checks don't catch sub-NaN corruption |
Which systems are affected
- Training on a node with a marginally faulty GPU
- Long pretraining runs
- Mixed precision (FP16/BF16/FP8)
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 settles above baseline with no NaN events
- ✓Verified signal present: Parameters drift from a reference run
- ✓Verified signal present: Validation metrics quietly degrade
- ✓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
- Hardware fault flips bits in computation without triggering NaN/Inf
- Errors accumulate as parameter drift, not crashes
- NaN checks don't catch sub-NaN corruption
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.