Azure OpenAI Chat with History

Back Next

Step 4: Configure and publish the processes

Configure the deployment details of the LLM model and publish the processes.

Open the

Read History from File

process.

On the

Start

tab of the

Start

step, select the same Secure Agent that you had selected in the

Run On

field for the

FileConnectionChatHistorydefault

connection.

Save and publish the process.

Open the

Write Chat History in File

process.

Optionally, in the

Prepare History to save in File

step, click the

Assignments

tab. Open the Expression Editor for the

File_Name

field and enter the format to save the file.

The following image shows the assignments of the

File_Name

input field:

Save and publish the process.

Open the

Chat with History

parent process.

On the Temp Fields
tab of the

Start

step, enter values for the following fields:

In the api_version
field, enter the API version of the LLM model. Default is

2024-06-01

. You can optionally edit the api version.

In the deployment_id
field, enter the user-specific deployment ID.

Optionally, in the

Configure Request Parameters

step, click the

Assignments

tab. Open the Expression Editor for the

Prompt_Configuration

field and enter the prompt instructions as shown in the following sample code:


<generationConfig_AzureAI>
      <max_tokens>200</max_tokens>
      <temperature>1</temperature>
      <topP>1</topP>
</generationConfig_AzureAI>

For the

Prompt_Configuration

field, enter values for the following properties:

Property	Description
maxTokens	Defines the maximum number of tokens that the model can generate in its response. Setting a limit ensures that the response is concise and fits within the desired length constraints.
temperature	Controls the randomness of the model's output. A lower value makes the output more deterministic, while a higher value increases randomness and creativity. For example, a temperature of 0.9 balances between deterministic and creative outputs.
topP	An alternative to sampling with temperature where the model considers the results of the token with topP probability. For example, if topP is set to 0.1, the model considers only the top 10% most probable tokens at each step.