1
Create a new dataset
Create a new dataset in your LangWatch workspace.
2
Add a traces column
Add a traces column with the JSON data type to store the trace data associated with each thread ID.

3
Add threads to the dataset
By selecting the thread mapping option, you can choose which information from the trace to include in the dataset. By default, the JSON object contains INPUT and OUTPUT fields.

4
That's it!
You can select multiple traces to add to the dataset. When using the thread mapping option, all traces will be grouped by the thread ID.