1Z0-1127-25 Using OCI Generative AI Service — 10 Questions

Question 1

A company uses OCI Generative AI Service to build a chatbot for customer support. They notice that the model sometimes generates inappropriate responses. What is the MOST effective way to mitigate this without retraining the model?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Configure system instructions to define acceptable behavior. Configuring system instructions is the most effective approach because it allows you to define the model's behavior and constraints at inference time without modifying the underlying model weights. In OCI Generative AI Service, system instructions act as a persistent prompt that guides the model's responses, enabling you to explicitly prohibit inappropriate content and enforce safety guidelines. This is a non-invasive, immediate mitigation that does not require the time, cost, or data preparation associated with retraining or fine-tuning.

Answer

Fine-tune the model with curated safe examples

Answer

Reduce the temperature parameter to 0

Answer

Use the moderation API to filter responses

Question 2

A developer wants to use OCI Generative AI Service to summarize long documents. Which endpoint should they use to send the document content?

Accepted Answer

/chat. The /chat endpoint in OCI Generative AI Service is designed for conversational interactions and can handle long document summarization by accepting the document content as part of the chat context. This endpoint supports multi-turn dialogues and large input payloads, making it suitable for processing and summarizing lengthy documents.

Answer

/generate

Answer

/classify

Answer

/embed

Question 3

An organization deploys a fine-tuned model for legal document analysis using OCI Generative AI Service. They need to ensure that only authorized users in the 'LegalTeam' group can access the model endpoint. Which policy statement should be used?

Accepted Answer

Allow group LegalTeam to use generative-ai-model in compartment ABC. The 'use' verb on the 'generative-ai-model' resource type grants the LegalTeam group permission to invoke the model endpoint for inference, which is the minimum privilege required for accessing a deployed fine-tuned model in OCI Generative AI Service. The 'use' permission specifically allows calling the model for text generation or analysis without granting broader management or read capabilities.

Answer

Allow group LegalTeam to manage generative-ai-family in compartment ABC

Answer

Allow group LegalTeam to read generative-ai-model in compartment ABC

Answer

Allow group LegalTeam to inspect generative-ai-model in compartment ABC

Question 4

A developer is using OCI Generative AI Service to generate code snippets. They want to ensure the output is as deterministic as possible for testing. Which combination of parameters should they use?

Accepted Answer

Temperature = 0, Top-p = 1. Setting Temperature=0 makes the model deterministic by always selecting the highest-probability token, while Top-p=1 includes all tokens in the sampling pool, ensuring no additional randomness is introduced. This combination eliminates stochastic variation, making outputs repeatable for testing.

Answer

Temperature = 0.5, Top-p = 0.5

Answer

Temperature = 0, Top-p = 0

Answer

Temperature = 1, Top-p = 1

Question 5

A company uses OCI Generative AI Service to build a chatbot for customer support. They notice that the model sometimes generates inappropriate responses. What is the MOST effective way to mitigate this without retraining the model?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Enable data redaction in the service. OCI Generative AI Service provides a built-in data redaction feature that automatically detects and removes personally identifiable information (PII) from training data before it is used for model training. This ensures compliance with data privacy regulations without requiring manual preprocessing or external tools.

Answer

Encrypt the training data with OCI Vault

Answer

Use the moderation API to scan outputs

Answer

Use a dedicated model endpoint

Question 6

Which TWO of the following are valid ways to reduce latency when using OCI Generative AI Service?

Accepted Answer

Use a dedicated AI cluster. Option A is correct because OCI Generative AI dedicated AI clusters provide reserved, isolated compute capacity for hosting models, so inference requests avoid contention on shared, multi-tenant endpoints and therefore experience lower, more predictable latency. Option B is correct because the max tokens parameter caps the number of tokens generated in the response; fewer generated tokens means less autoregressive decoding work, which directly shortens response time and thus reduces latency. Option C is not a valid general answer because moving to another region does not inherently lower latency and may even increase it depending on network distance to the client. Option D is wrong because larger models generally have more parameters and higher inference cost, which tends to increase rather than reduce latency. Option E is wrong because batching multiple requests increases per-request queueing and processing time, which raises latency rather than reducing it.

Answer

Deploy the model in a different region

Answer

Use a larger model

Answer

Batch multiple requests

Question 7

Which THREE of the following are supported capabilities of OCI Generative AI Service?

Accepted Answer

Text summarization. OCI Generative AI Service supports text summarization (A), which condenses long documents into concise summaries using its hosted large language models. It also supports question answering (D), allowing users to query a model or a dedicated endpoint backed by a knowledge base to retrieve answers from provided content. Code generation (E) is likewise supported, as the service's models can produce, complete, and explain code in multiple programming languages. Sentiment analysis (B) is not a core generative capability of the service; it is typically handled by OCI Language, a separate AI service. Image generation (C) is not offered by OCI Generative AI Service, which focuses on text and code generation rather than image synthesis.

Answer

Sentiment analysis

Answer

Image generation

Question 8

Which TWO of the following are required to fine-tune a model using OCI Generative AI Service?

Accepted Answer

A training dataset in the required format. Option A is correct because fine-tuning in OCI Generative AI Service requires a training dataset uploaded to Object Storage in the JSONL format expected by the service, with each line containing prompt/completion pairs (or chat-style messages) that the tuning job consumes. Option B is correct because the fine-tuning job must reference the base model identifier (for example, a specific pretrained model such as cohere.command or meta.llama-3 variants) so the service knows which foundation model to adapt with the custom data. Option C is not required because quota and compartment are handled by IAM policies and service limits, not as a distinct fine-tuning input parameter. Option D is not required because authentication is handled through OCI IAM credentials or instance principals, not a separate API key specific to fine-tuning. Option E is not required because OCI Generative AI fine-tuning is a managed service and does not require the customer to provision a dedicated AI cluster.

Answer

A compartment with sufficient quota

Answer

An OCI API key

Answer

A dedicated AI cluster

Question 9

Refer to the exhibit. A developer sends this JSON payload to the /chat endpoint. The response includes an error that 'maxTokens' must be an integer. What is the issue?

Accepted Answer

The parameter should be 'max_tokens' instead of 'maxTokens'. The OCI Generative AI service expects the parameter name 'max_tokens' (snake_case) for specifying the maximum number of tokens in the response, not 'maxTokens' (camelCase). The error message indicates that the value is not being recognized as an integer because the JSON key itself is incorrect, causing the service to fail validation.

Answer

The compartmentId is missing

Answer

The temperature value is too low

Answer

The model name 'cohere.command-light' is incorrect

Question 10

A startup is building a chatbot for customer support using OCI Generative AI Service. The chatbot needs to answer queries about product features based on a knowledge base of product documentation. Which configuration is most appropriate for this use case?

Accepted Answer

Use a Cohere Command model with the knowledge base as context in a prompt, and enable retrieval-augmented generation (RAG) via OCI Generative AI Agents.. OCI Generative AI Agents with retrieval-augmented generation (RAG) allows the chatbot to dynamically retrieve relevant chunks from the product documentation knowledge base and inject them as context into a Cohere Command model prompt. This approach ensures answers are grounded in the latest documentation without requiring fine-tuning, and it scales efficiently as the knowledge base grows.

Answer

Use the Summarization task type to generate concise answers from the documentation.

Answer

Fine-tune a Llama 2 70B model on the product documentation to create a custom model.

Answer

Use the Code Generation model to produce SQL queries that retrieve answers from a database.