What does this failure look like in production?

In production this failure typically appears during the first tens-to-hundreds of training steps: loss curves become non-monotonic, validation metrics plateau above the baseline, and downstream checkpoints either diverge or fail to improve. Engineers frequently misread it as data drift when the underlying mechanism is fp16 has limited exponent range causing overflow. Validating against the symptom list in this entry is the fastest way to confirm the diagnosis.

Is this a hardware or a software issue?

Training Stability failures can look like either layer depending on the symptom. The discriminator is the resolution of the symptom to the recommended fix: if applying the steps under "How to fix it" eliminates the symptom, this is a software/configuration class issue regardless of the hardware it surfaced on. Repeat the fix on a fresh node to confirm.

How long does the fix typically take?

Most resolutions from this entry take 10 to 30 minutes of engineer time once the diagnosis is confirmed. Long fixes come from skipping the diagnosis step and chasing apparent symptoms (validation drift, memory growth, slow throughput) instead of the actual root cause listed above. Apply the fix in a small experiment first, then roll it forward.

Does this affect inference or only training?

Primarily this is a training stability failure observed during training, but the same misconfiguration can leak into downstream evaluation and serving. Treat the prevention checklist as a guardrail: any pipeline that touches a trained checkpoint should re-validate the failure mode before promoting the model.

What's the cheapest way to confirm this is the problem?

Cheapest signal: reproduce the symptom on a 1-GPU toy run with the same configuration. If the toy run exhibits the same behavior, the cause is the configuration rather than the cluster. If only the multi-GPU run is affected, the failure is a distributed interaction, Training with torch.cuda.amp is the most likely layer to inspect.

Can the Denpex platform detect this automatically?

Yes. Denpex correlates the symptoms in this entry against your training logs in real time and surfaces the failure taxonomy, root cause, and exact remediation command on the console. Three full diagnoses are included free during signup, and the encyclopedia entry you are reading is generated from the same signal library that powers the platform.

What related failures should I check next?

Closely related failures cluster around #bf16, #fp16, #mixed-precision. After applying this fix, validate against the related entries below to confirm no overlapping cause is masking as this one. Engineers routinely discover that the dominant failure mode is actually a second, sibling failure compounded on top.

Will this recur after the fix?

Applied once with the prevention checklist, this failure does not recur under normal operation. The recurring variant most engineers encounter is the re-emergence of the same symptom on a different model or stack after bf16 is upgraded. Pin compatible versions and rerun the prevention smoke test on every dependency change.

How does this scale from 1 GPU to 1000 GPUs?

At single-GPU scale the symptom usually looks like slow convergence or a single anomalous loss value. At 64+ GPU scale, distributed interactions amplify fp16 has limited exponent range causing overflow through collectives, and the failure can manifest as a full job abort. Always reproduce the fix at the smallest cluster configuration that still exhibits the symptom before promoting to production scale.

Is there a fingerprint I can monitor?

Yes. FP16 NaN with certain operations is the highest-signal fingerprint. Wire a metric or alert on that signal (or its absence after the fix) so you can detect the recurrence automatically. The Denpex platform encodes this exact fingerprint and surfaces it directly in the dashboard.

What is the official documentation that covers this?

Reference documentation is listed in the References section at the bottom of this entry. The most authoritative URLs are the upstream framework docs and the issue trackers cited; the Denpex entry synthesizes both into the executive, engineer, and on-call perspectives so each audience can act without reading every linked source.

Does this entry include working code?

Yes. The "Code Examples" section contains copy-paste-ready snippets that target the most common stacks (PyTorch, Hugging Face Trainer, DeepSpeed, Megatron-LM, vLLM/Transformers). Each snippet is annotated with why it works so you can adapt the pattern to your own framework without losing fidelity.

AMP BF16 vs FP16 Confusion

BF16 and FP16 mixed precision have different numerical properties; choosing the wrong one causes instability or wasted memory.

Quick answer

BF16 and FP16 mixed precision have different numerical properties.

Root cause: FP16 has limited exponent range causing overflow. BF16 has same exponent as FP32 but less precision. FP16 needs loss scaling, BF16 doesn't.
Recommended fix: Use BF16 on H100, A100, and modern GPUs (Ampere+)
How Denpex helps: Denpex investigates AMP BF16 vs FP16 Confusion using the evidence you provide or your connected workload collects. Earlier rank, host or application evidence is needed to distinguish an initiating failure from a downstream report.

Training Stability#bf16#fp16#mixed-precision#gradscaler#training-stability

What this failure is

AMP BF16 vs FP16 Confusion is a Training Stability failure seen during ML training runs. BF16 and FP16 mixed precision have different numerical properties; choosing the wrong one causes instability or wasted memory. Common tags: Bf16, Fp16, Mixed Precision, Gradscaler.

Live diagnosis, no signup

Is this what broke your run? Paste your log.

You're reading about AMP BF16 vs FP16 Confusion. Paste your own traceback and relevant evidence for an investigation of your workload, with a next action or a specific missing fact. A reference entry does not establish your cause. No account or card for the free diagnosis. Review data handling before submitting sensitive logs.

Before uploading, review cloud data handling and local options.

training_logs.txt

No log to hand? Try one:

Upload or drop a plain-text log

Upload your log file. Large files are processed in your browser. Only selected evidence is submitted when you run diagnosis.

3 free diagnoses/day

Why it happens (the mechanism)

FP16 has limited exponent range causing overflow. BF16 has same exponent as FP32 but less precision. FP16 needs loss scaling, BF16 doesn't. BF16 is preferred for H100/A100. 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 unstable with FP16
FP16 overflow causes NaN
BF16 uses more memory than expected

Common symptoms and what they mean

Symptom	Why it happens
FP16 NaN with certain operations	FP16 has limited exponent range causing overflow
BF16 model is less accurate than FP32	BF16 has same exponent as FP32 but less precision
Mixed precision doesn't speed up training	FP16 needs loss scaling, BF16 doesn't

Which systems are affected

Training with torch.cuda.amp
Mixed precision training
Using H100/A100 with BF16

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: FP16 NaN with certain operations
✓Verified signal present: BF16 model is less accurate than FP32
✓Verified signal present: Mixed precision doesn't speed up training
✓A targeted fix from the "How to fix it" section eliminates or substantially reduces the symptom within one validation pass.

Root cause

FP16 has limited exponent range causing overflow
BF16 has same exponent as FP32 but less precision
FP16 needs loss scaling, BF16 doesn't
BF16 is preferred for H100/A100

The fix and how to prevent it

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.

No 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 extension