최신Snowflake SnowPro® Specialty: Gen AI Certification - GES-C01무료샘플문제

A.

B.

C.

D.

E.

A. The exported SQL query, when used with dynamic tables, supports incremental refresh for efficient processing of new data without recomputing the entire table.

B. The Playground allows exporting the exact SQL query with all defined model settings, including temperature and Cortex Guard enablement, for direct use in a Snowflake worksheet or task.

C. To filter unsafe LLM responses in production, the Cortex Guard option, which is built with Meta's Llama Guard 3, must be explicitly enabled in the COMPLETE function's options argument.

D. For continuous processing of new data, the exported SQL query can be automated using

E. If the mistral-large2 model is not natively available in the target production region, cross-region inference must be enabled by setting the CORTEX_ENABLED_CROSS REGION parameter.

A. Deploying a custom fine-tuned model using SNOWFLAKE.CORTEX.FINETUNE on proprietary documentation is the most effective approach to ensure factual accuracy for any LLM task.

B. For tasks requiring LLMs to generate SQL queries from natural language, using the Cortex Analyst verified Query Repository (VQR) can improve accuracy by leveraging pre-verified SQL queries for similar questions.

C. Using Cortex Search as a Retrieval Augmented Generation (RAG) engine can enhance LLM responses by providing relevant context from proprietary documentation, thereby reducing hallucinations.

D. AI Observability can be leveraged to systematically evaluate applications, measuring metrics like 'factual correctness and 'groundedness' to detect and mitigate hallucinations, especially in summarization.

E. Enabling Cortex Guard with guardrails: true directly addresses model hallucinations by ensuring responses are always factually correct and aligned with the provided context.

A. Option D

B. Option E

C. Option C

D. Option A

E. Option B

A. Real-time LLM inference for a conversational interface is ideally deployed as a ' 'Service' ' in SPCS, which is long-running and accessible via an HTTP endpoint, ensuring continuous availability and responsiveness.

B. When deploying LLMs to SPCS, it's generally most cost-efficient to use generic CPU instance types like 'CPU X64 XS' for all tasks, as GPU instances (e.g., are exclusively for highly specialized computer vision tasks and not optimized for LLMs.

C. Container images for SPCS deployments are typically pushed to a public Docker Hub repository, and Snowflake pulls them as needed during service creation and scaling, simplifying image management.

D. GPU-accelerated LLM training, which is a finite and often resource-intensive task, is best implemented as a ''job'' in SPCS, invoked via "EXECUTE JOB SERVICE', as it is designed to run to completion and then spin down.

E. For batch inference on Snowflake data where data locality and efficiency are key, using ''Service Functions'' is highly efficient because data is passed as input parameters directly from SQL queries, and this design ensures the data never leaves the Snowflake network boundary.

A.

B.

C.

D.

E.

A.

B.

C.

D.

E.

A. Any internal stages used for storing documents that will be processed by Document AI must explicitly enable SNOWFLAKE_SSE encryption.

B. The database and schema where Document AI model builds are created can be freely altered after creation, allowing for agile schema evolution.

C. All documents intended for a single Document AI '!PREDICT operation must be stored within the same logical directory of the specified stage.

D. Snowflake recommends using a large virtual warehouse, such as an 'L' or 'XL' size, to accommodate the intensive processing demands of Document AI and ensure high throughput.

E. A dedicated, smaller warehouse (e.g., 'X-SMALL', 'SMALL', or 'MEDIUM') should be created for Document AI to facilitate precise cost tracking, as scaling up warehouse size does not enhance Document AI query performance.

A.

B.

C.

D.

E.

A. The 'VECTOR data type is not supported in 'VARIANT columns, which means direct storage of embeddings alongside other semi-structured metadata in a single "VARIANT column is not possible.

B. When 'EMBED_TEXT 1024' is invoked within a Snowflake dynamic table's SELECT statement, it allows for continuous, automated updates of embeddings as new data arrives.

C. The 'VECTOR data type, used to store the output of is fully supported as primary or secondary index keys in Snowflake's hybrid tables.

D. The 'EMBED function in the Cortex REST API can be used to process a list of text strings, where each individual string can be up to 4096 characters long.

E. To compare the generated 1024-dimension vectors for similarity, only the 'VECTOR COSINE SIMILARITY function is officially supported by Snowpark Python.

A. Detailed request and response bodies, along with the generated SQL, are stored and can be directly queried using standard SQL.

B. Logs are exclusively available for analysis through pre-built dashboards in Snowsight and cannot be accessed via direct SQL queries.

C. The logs are automatically purged after 7 days of being recorded, requiring a separate process for long-term data retention.

D. Access to these detailed event tables is implicitly granted to roles holding the SNOWFLAKE. CORTEX_USER database role and the AI_OBSERVABILITY_EVENTS_LOOKUP application role.

E. Logged data from AI Observability's event tables becomes visible within a small latency, typically 1-2 minutes, after a request is made.

A. The 'GET_PRESIGNED_URL' function used in the '!PREDICT query has an expired URL.

B. The PDF documents exceed the maximum allowed file size of 50 MB.

C. The Document AI model build is attempting to process more than 1000 documents in a single query.

D. The documents contain non-English text, which is not fully supported by Document AI for optimal results.

E. The internal stage was not created with 'SNOWFLAKE_SSE' encryption enabled.