# NaN and Inf Detection

NaN and Inf counts are computed as standard [tile statistics](/docs/inspect-and-analyze/tensor-statistics) for every tensor Tensormorph indexes - not a special diagnostic mode you turn on, but always-available numbers that happen to be exactly what you need when a training run has gone wrong.

## Why this needs to be structural, not a scan

A naive "scan every value looking for NaN" approach doesn't work at the scale Tensormorph targets - checking a trillion-parameter checkpoint element-by-element on demand would be far too slow to be useful interactively. Because NaN/Inf counts are computed once during indexing as part of the same mergeable tile-statistics pass as everything else, "does this model have any NaNs, and where" is answered by reading already-computed counts, aggregated up from tile to layer to model, not by re-scanning anything.

## Finding the source of a corruption

Because the counts are mergeable, locating a NaN is a top-down narrowing process rather than a flat search:

1.  At the model level, the aggregate NaN/Inf count tells you immediately whether anything is wrong.
2.  Drilling into the [Architecture view](/docs/editors/architecture-editor), layers and modules with a non-zero count are visually distinguishable (a dedicated overlay/shading option highlights them directly), so you can navigate straight to the affected region instead of reading through the whole tree.
3.  At the tile and scalar level, the [Scalar Inspector](/docs/editors/scalar-inspector) shows the exact corrupted values.

## Querying for corruption

The same counts are queryable through [TQL](/docs/reference/expressions-and-queries) - a query like "every tensor with a non-zero NaN count" runs as a filter over precomputed statistics across the whole model, immediately, rather than a live scan. This is also the basis for a **breakpoint** condition in [runtime tracing](/docs/profiling/runtime-traces): a capture point can be set to trigger specifically when a NaN first appears in a given tensor during a forward or backward pass, which is usually a far more useful place to stop than an arbitrary step number.

## Related resources

-   [Tensor Statistics](/docs/inspect-and-analyze/tensor-statistics) - where NaN/Inf counts sit among the other precomputed statistics.
-   [Runtime Traces](/docs/profiling/runtime-traces) - using NaN detection as a breakpoint condition during a live run.
-   [Outliers and Anomalies](/docs/inspect-and-analyze/outliers-and-anomalies) - for values that are finite but still suspicious.