1.3 Explain Role of Context Windows, Token Limits and Response Management AI-TECHNICAL-PRACTITIONER Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   What is the purpose of truncating a model's response when it reaches the token limit?

   • ATo prevent the response from exceeding the maximum allowed tokens
   • BTo improve the model's creativity
   • CTo reduce the input prompt size
   • DTo ensure the response is complete and accurate
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A developer is building an application that uses a language model to answer questions about a large knowledge base. The developer notices that when the user asks a follow-up question, the model forgets the earlier part of the conversation. What is the most likely cause?

   • AThe model's temperature setting is too high, causing it to ignore previous messages.
   • BThe model is not designed to handle multi-turn conversations.
   • CThe model's context window is being filled with the entire conversation history, causing earlier messages to be dropped.
   • DThe developer is not using a large enough max_tokens parameter for the response.
   Reveal Solution Discussion Select an answer first
4. 8expert · hard

   A company is using a language model to answer questions about a large codebase. The codebase is 20,000 tokens, and the model's context window is 8,000 tokens. The team wants to provide accurate answers about any part of the codebase. Which approach is most effective?

   • AIndex the codebase into chunks and retrieve the most relevant chunks for each question.
   • BSend the entire codebase and rely on the model to find the relevant code.
   • CUse a model with a larger context window, but only if it is available.
   • DSummarize the entire codebase into a single high-level description and use that for all questions.
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A team is using a language model to generate a summary of a 6,000-token report. The model's context window is 4,000 tokens. The team wants the summary to be comprehensive and include information from all sections. Which approach is most effective?

   • ATruncate the report to the first 4,000 tokens and summarize that portion.
   • BUse a higher temperature to make the model process more tokens.
   • CSplit the report into two chunks, summarize each chunk, and then combine the summaries into a final summary.
   • DSend the entire report and ask the model to output a summary of exactly 4,000 tokens.
   Reveal Solution Discussion Select an answer first
6. 10expert · hard

   A team is using a language model to analyze a 7,000-token research paper. The model's context window is 5,000 tokens. The team needs to extract key findings from the entire paper. Which strategy is most effective?

   • ATruncate the paper to the first 5,000 tokens and extract findings from that portion.
   • BUse a higher temperature to make the model process more tokens.
   • CSend the entire paper and ask the model to extract findings from the beginning.
   • DSplit the paper into two parts, extract findings from each part, and then combine the findings.
   Reveal Solution Discussion Select an answer first