How do I configure the Basic Language Learning model node?#
A Chain node extends the functionality of a connected Large Language Model or Chat Model by specifying how inputs are handled from end-users. The internal logic is completely customizable, as opposed to [Agent nodes which automatically determine the correct node for a given user output.
Use the Basic Language Learning model to set: * a prompt for a chosen model, and * an optional parser to process the response
Before you begin#
Step 1 - add the node#
- Click + to open the nodes panel
- Search for Basic Language Learning model in the node search field.
- Click the node to add it to the canvas.
Step 2 - define how the node receives input data#
Note
Click Test to load values from input nodes you can drag and drop into fields to map to your output values
Configure the node to receive input data:
| Prompt | Description | Additional information |
|---|---|---|
| Take from previous node automatically | Node assumes input field from previous node is defined as chatInput |
|
| Define below | Define an expression to gather input or add static text | Learn how to add expressions to a node |
Step 3 - Define output format (optional)#
Toggle Define specific output format to connect an output parser to the node.
Step 4 - add chat model prompt (optional)#
Choose from three prompts to send to an attached Chat model:
| Prompt | Description | Additional information |
|---|---|---|
| AI | Enter a given response that the chat model will attempt to emulate | |
| System | Enter a system message that the chat model will emulate regardless of end-user input | |
| User | Add a suggested structure for end-users to follow when entering their questions. The structure and end-user input is used by the model to emulate in responses |