> For the complete documentation index, see [llms.txt](https://docs.radar.citadel-ai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.radar.citadel-ai.com/en/application-owner-guide/monitor-applications/traffic.md).
# Traffic
On the Traffic page, you can view a list of input and output logs for your application.
Each log displays Evaluation Metric results, `Block` / `Flag` status, and detected Evaluation Categories. Use this page to inspect individual logs and investigate detection details.
### Open the Traffic page
The Traffic page can be opened from the application details page.
On the `Nodes` page, select the target application, then select the `Traffic` card on the application details page.
### Filter logs
On the traffic page, you can specify conditions such as `request` / `response`, `Block` / `Flag`, and period to filter the logs you want to review.
{% hint style="info" %}
**Note: request and response logs**
The Traffic page separates logs into input evaluation and output evaluation.
Choose Requests if you want to inspect input metric detections, or Responses if you want to inspect output metric detections.
Although output evaluations may assess both user inputs and LLM outputs together, input and output evaluation logs are stored as separate entries.
This is because input and output evaluations serve different analytical perspectives. Additionally, if an input is blocked during evaluation, no corresponding output log will exist. This design also avoids requiring clients to implement extra handling to pair requests and responses into a single session.
Therefore, Citadel Radar manages requests and responses separately, allowing you to inspect each log according to the focus of your investigation.
{% endhint %}
If you want to filter in more detail, use advanced filtering. Advanced filtering lets you extract logs by combining conditions such as borderline handling, metadata, and Evaluation Metrics.
{% hint style="info" %}
If you want to filter using metadata, metadata must be included in the API Request Body. For details, see the [API endpoints](/en/developer-guide/api-endpoints.md) page.
{% endhint %}
### Review log details
In the log list, you can check the evaluation results and detected Evaluation Categories for each log. If you want to view more details for a specific log, open its details from `Actions`.
In the details, you can review input content, output content, evaluation results for each Evaluation Metric, metadata, and more.
Additionally, you can use `AI Explanation` to generate breakdowns showing exactly which parts of a log were flagged as problematic. This helps you review detection findings and determine appropriate response strategies.
You can also add comments to logs. Recording investigation details and response policies helps with team review and follow-up actions.