> ## Documentation Index
> Fetch the complete documentation index at: https://docs.axioniclabs.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Observability

> Review API request activity, inspect prompts and responses, and check SAE behavior-monitoring readiness for every model you operate in Spectra.

Observability combines request activity and behavior-monitor readiness in one route so you can validate both runtime traffic and monitor setup from the same workspace.

## Request activity

Review recent inference calls, including:

* route and request status
* model used
* prompt snippet
* timestamps and recent history

Use the search bar to filter by route, model, or prompt text. You can also switch between the full activity feed and SAE-focused activity when you only want monitored requests.

Selecting a row opens the call details panel, where you can inspect the prompt, output, and request parameters in more detail.

## Behavior readiness

The readiness summary shows which behaviors are ready for SAE-backed runtime monitoring and which ones still need setup work.

Use it to separate behaviors into:

* fully ready for monitored generation
* partially configured and still missing required pieces
* the full behavior list across the workspace

In practice, a behavior is only monitor-ready once both its steering vector link and its SAE baseline are present.

## How it fits with the rest of Spectra

* create rules in [Behaviors](/products/spectra/behaviors)
* apply and test them in [Optimization](/products/spectra/optimization)
* use Observability to confirm the monitoring layer is configured and actually being exercised