1z0-1127-24 Sample Questions Answers

The Oracle Accelerated Data Science (ADS) class that can be used to deploy a Large Language Model (LLM) application to OCI Data Science model deployment is GenerativeAI. This class provides the necessary tools and functions to work with generative AI models, including deployment, fine-tuning, and inference capabilities. It integrates with OCI Data Science to streamline the process of deploying and managing LLM applications.

References

Oracle ADS documentation

Guides on deploying AI models using Oracle Data Science services

Chain-of-Thought prompting involves prompting the Large Language Model (LLM) to emit intermediate reasoning steps as part of its response. This technique helps the model articulate its thought process and reasoning, leading to more transparent and understandable outputs. By breaking down the problem into smaller, logical steps, the model can provide more accurate and detailed responses.

References

Research articles on Chain-of-Thought prompting

Technical guides on enhancing model transparency and reasoning with intermediate steps

In Retrieval-Augmented Generation (RAG), two key evaluation metrics are Groundedness and Answer Relevance, each addressing different aspects of AI-generated responses.

Groundedness (Factual Correctness):

Determines whether the AI-generated answer is factually based on the retrieved data.

Ensures that AI responses are verifiable from the reference documents.

Prevents hallucination, where the model generates plausible but incorrect answers.

Answer Relevance (Query Alignment):

Evaluates how well the response aligns with the user’s original query.

A response may be factually correct but irrelevant to the actual question.

Ensures that AI responses remain contextually appropriate.

Key Differences:

Groundedness ensures that responses are based on correct data.

Answer Relevance checks whether the response actually answers the user’s query.

???? Oracle Generative AI Reference:Oracle AI ensures fact-checking and contextual accuracy in its retrieval-augmented models, reducing AI hallucination risks.

The "temperature" parameter in generative AI models controls the randomness of the model's output. It affects the creativity and diversity of the generated text:

Low temperature: Leads to more deterministic and focused outputs, where the model tends to choose the most probable tokens, resulting in less randomness and creativity.

High temperature: Increases randomness by making the probability distribution over the next tokens flatter. This allows for more diverse and creative outputs, as the model is more likely to choose less probable tokens.

Adjusting the temperature parameter enables fine-tuning the balance between creativity and coherence in the model's responses.

References

Research articles on the role of temperature in generative models

Technical guides for tuning generative AI models in OCI

StreamlitChatMessageHistory is a chat message storage tool in Streamlit, used to manage message history within LLM-powered applications.

Key Features of StreamlitChatMessageHistory:

Stores chat messages within Streamlit’s session state.

Not persistent across sessions; resets when the session is closed.

Specific to Streamlit applications, not designed for all LLM applications.

Why Option (D) is Incorrect:

StreamlitChatMessageHistory is designed for Streamlit-based apps.

It is not suitable for all LLM applications, particularly those requiring persistent storage.

Why Other Options Are Correct:

(A) True: Each session has its own instance of StreamlitChatMessageHistory.

(B) True: It is not persisted across sessions.

(C) True: It stores messages in the Streamlit session state.

???? Oracle Generative AI Reference:While Oracle AI supports various LLM applications, StreamlitChatMessageHistory is limited to Streamlit-based chat interfaces.

In the inference workflow of the OCI Generative AI service, a "model endpoint" is a critical component. It serves as a designated point for handling user requests and providing model responses. When users or applications send requests to the model endpoint, the endpoint processes these requests by passing them to the deployed model. The model then generates responses based on the input data, and these responses are returned to the user through the same endpoint. This setup facilitates efficient and scalable interaction with the AI model, ensuring that inference can be performed seamlessly and reliably.

References

Oracle Cloud Infrastructure (OCI) Generative AI service documentation

General principles of model deployment and inference in cloud services

Vector indexing plays a crucial role in vector search and retrieval systems, particularly in AI-driven databases. The key functions of vector indexing include:

Efficient Search and Retrieval – Vector indexing structures (such as HNSW, FAISS, or Annoy) help organize vector embeddings to enable fast retrieval of similar vectors.

Mapping to Searchable Data Structures – The process involves creating indexes that efficiently store and map vectors, reducing computational overhead when searching for similar embeddings.

Handling High-Dimensional Data – Since vector embeddings (used in NLP, image recognition, etc.) are often high-dimensional, indexing helps compress and cluster similar vectors, improving retrieval speed.

Used in Vector Databases – Many AI applications, including Oracle's AI-driven database solutions, use indexing techniques for faster similarity searches.

???? Oracle Generative AI Reference:Oracle integrates vector search within its AI and database services, allowing enterprises to efficiently manage and retrieve vectorized data.

In the "Show Likelihoods" feature of language model token generation, a higher number assigned to a token indicates that the token is more likely to follow the current token. This likelihood is based on the model's probability distribution, where tokens with higher probabilities are considered more likely to be the next in the sequence. This feature helps in understanding the model's decision-making process and the relative probabilities of different tokens.

References

Technical documentation on language model token generation

Research articles on probability distributions in generative models

LangSmith Evaluators are not typically used for aligning code readability. Instead, they are used for tasks such as measuring the coherence of generated text, evaluating the factual accuracy of outputs, and detecting bias or toxicity. Evaluators help ensure the quality and reliability of the outputs generated by language models.

References

LangSmith documentation on evaluators

Research articles on evaluation metrics for language models

The architecture of dedicated AI clusters contributes to minimizing GPU memory overhead for fine-tuned model inference by sharing base model weights across multiple fine-tuned models on the same group of GPUs. This approach allows different fine-tuned models to leverage the shared base model weights, reducing the memory requirements and enabling efficient use of GPU resources. By not duplicating the base model weights for each fine-tuned model, the system can handle more models simultaneously with lower memory overhead.

References

Technical documentation on AI cluster architectures

Research articles on optimizing GPU memory utilization in model inference

K-shot prompting refers to providing the language model with k examples of the task at hand within the prompt. These examples help guide the model's understanding and output by demonstrating the desired format and approach. This technique is used to improve the model's performance on specific tasks by showing it how to handle similar situations.

References

Research papers on few-shot learning and prompting techniques

Technical documentation on using examples in prompts for large language models

Using small data sets with the Vanilla fine-tuning method in the OCI Generative AI service might result in underfitting. Underfitting occurs when a model is too simplistic to capture the underlying patterns in the data, leading to poor performance on both training and validation data. This is particularly problematic with small data sets because there may not be enough information for the model to learn the necessary patterns and relationships.

References

Articles on machine learning challenges with small data sets

Technical documentation on fine-tuning models in OCI

In the simplest form of keyword-based search, documents are evaluated based on keyword matching and term frequency. This approach does not account for context, semantics, or the meaning behind the words, but rather focuses on:

Presence of Keywords – If a document contains the search term, it is considered relevant.

Term Frequency (TF) – The more a keyword appears in a document, the higher the ranking in basic search algorithms.

Inverse Document Frequency (IDF) – Words that are common across many documents (e.g., “the,” “is”) are given less weight, while rare words are prioritized.

Boolean Matching – Some basic search engines support logical operators like AND, OR, and NOT to refine keyword searches.

Exact Match vs. Partial Match – Some systems prioritize exact keyword matches, while others allow partial or fuzzy matches.

???? Oracle Generative AI Reference:Oracle has implemented semantic search and advanced AI-driven document search techniques in its cloud solutions, but traditional keyword-based search still forms the foundation of many enterprise search mechanisms.

The "stop sequence" parameter in the OCI Generative AI Generation models is used to specify a string that signals the model to stop generating further content. When the model encounters this string during the generation process, it terminates the response. This parameter is useful for controlling the length and content of the generated text, ensuring that the output meets specific requirements or constraints.

References

OCI Generative AI service documentation

General principles of sequence generation in AI models

Cosine distance (or cosine similarity) is a metric used to measure the angular similarity between two vectors in high-dimensional space.

Cosine Distance Calculation:

Cosine similarity formula:

A black and white background with black text AI-generated content may be incorrect.

The value ranges from -1 to 1:

1 → Vectors are identical.

0 → Vectors are orthogonal (unrelated).

-1 → Vectors are completely opposite.

Why a Cosine Distance of 0 Means Similar Direction:

A cosine similarity of 1 means vectors point in the same direction.

A cosine distance of 0 means maximum similarity (no angular difference).

Why Other Options Are Incorrect:

(A) is incorrect because a cosine distance of 0 implies similarity, not dissimilarity.

(B) is incorrect because unrelated vectors have a cosine similarity close to 0, not exactly 0.

(C) is incorrect because cosine similarity does not measure vector magnitude, only direction.

???? Oracle Generative AI Reference:Oracle's vector search and embedding-based AI models rely on cosine similarity for semantic search, recommendation systems, and NLP tasks.

In LangChain, the "mmr" (Maximal Marginal Relevance) search type is used to balance between relevancy and diversity when retrieving documents. This technique aims to select documents that are not only relevant to the query but also diverse from each other. This helps in avoiding redundancy and ensures that the retrieved set of documents covers a broader aspect of the topic.

Maximal Marginal Relevance (MMR) works by iteratively selecting documents that have high relevance to the query but low similarity to the documents already selected. This ensures that each new document adds new information and perspectives, rather than repeating what is already included.

References

LangChain documentation on retrievers and search types

Research papers and articles on Maximal Marginal Relevance (MMR)