AI-102 exam dumps

AI-102 practice question 492 of 493

Designing and Implementing a Microsoft Azure AI Solution. Professional level, Microsoft. Free question with the correct answer and a full explanation.

AI-102 Question 492

Select 3

You are tasked with fine-tuning an Azure OpenAI model to improve its performance for a customer support chatbot that handles domain-specific queries. Which of the following steps are required to fine-tune the model successfully in Azure OpenAI?

  1. A

    Prepare a dataset in JSONL format with training examples and validation examples.

  2. B

    Configure a custom endpoint URL for your fine-tuned model during the fine-tuning process.

  3. C

    Upload the training dataset to an Azure Blob Storage container accessible to the Azure OpenAI resource.

  4. D

    Use the Azure OpenAI Studio or Azure CLI to initiate the fine-tuning process with the prepared dataset.

  5. E

    Manually adjust the learning rate and epochs during the fine-tuning process.

Show answer and explanation

Correct answers: A, C, D

Explanation

Fine-tuning an Azure OpenAI model involves preparing a dataset in JSONL format, uploading the dataset to an accessible storage location, and initiating the fine-tuning process using tools like Azure OpenAI Studio or Azure CLI. Azure OpenAI handles the hyperparameters for fine-tuning, such as learning rate and epochs, automatically, and the endpoint URL for the fine-tuned model is generated by Azure.

  • A. Correct.

    Correct. The dataset must be in JSONL format, and it should include labeled training examples for fine-tuning the model. This is a required step.

  • B. Incorrect.

    Incorrect. The endpoint URL is automatically generated for the fine-tuned model. You do not configure it manually during the fine-tuning process.

  • C. Correct.

    Correct. The training dataset must be uploaded to Azure Blob Storage or another accessible location for the Azure OpenAI service to access it.

  • D. Correct.

    Correct. Fine-tuning in Azure OpenAI can be initiated using Azure OpenAI Studio, Azure CLI, or REST API with the prepared dataset.

  • E. Incorrect.

    Incorrect. Learning rate and epochs are pre-configured by Azure OpenAI during the fine-tuning process, and manual adjustments are not required.

Timed practice exam

Take a AI-102 practice test under exam conditions

40 questions in 120 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam