Practice context window vocabulary: token limits, document chunking, batch processing, attention degradation in long contexts, and context compression strategies.
0 / 10 completed
1 / 10
'The context window is 128K _____.' What unit measures the amount of text a model can process at once?
Tokens are the units LLMs process — roughly 3/4 of a word in English. A 128K token context window can hold approximately 100,000 words or ~300 pages of text.
2 / 10
'The document exceeds the _____ window.' What is the limit being exceeded?
When a document is too large for the model to process in one call, it 'exceeds the context window' — requiring chunking, summarization, or retrieval strategies.
3 / 10
'We _____ the document and process in batches.' What technique handles oversized documents?
'Chunking' means splitting a large document into smaller, overlapping pieces that fit within the context window, then processing each chunk separately.
4 / 10
'The model's attention degrades in the middle of a long context' — this is known as the _____ problem.
The 'lost in the middle' problem refers to research showing LLMs perform worse on information placed in the middle of long contexts vs. the beginning or end.
5 / 10
What does 'context compression' refer to?
Context compression strategies (summarization, selective retrieval, prompt pruning) maximize the useful information that fits within the context window.
6 / 10
Alex: 'I'm seeing some latency with the API call to the summarization service. The response time is consistently around 3 seconds.'
Which of the following best explains why this might be happening in relation to the context window size?
Alex's observation highlights a common issue when dealing with large context windows. Models often spend more time processing the entire input sequence before generating a response, especially if it exceeds the window size. The model might be re-evaluating information repeatedly within the context window, leading to delays. Option A is less directly related to the context window problem; options C and D propose alternative causes for latency.
7 / 10
Sarah (in a Slack channel): 'Just ran a quick test with the new prompt. The output is completely nonsensical - it's referencing details from earlier in the document that should have been irrelevant!'. What technical term best describes Sarah's observation regarding the model's behavior?
'Hallucination' is a well-established term describing a common issue in large language models – the generation of factually incorrect or irrelevant information. While 'context drift' describes changes to the context during processing, hallucination more directly represents Sarah's experience of the model pulling in unrelated details. 'Token decay' and 'semantic confusion' relate to different aspects of model performance but aren't the primary explanation for this specific scenario.
8 / 10
During a code review, Ben comments on a PR: 'I'm concerned about the length of this prompt. It's pushing close to the context window limit – we might experience performance degradation if it keeps growing.' Which technique would be most appropriate for Ben to implement to mitigate this risk?
A sliding window approach is a standard technique for managing large context windows. By processing only the most recent parts of the document – discarding older information – you keep the input size within the model's limits, preventing performance degradation. Options A, C, and D address different aspects of prompt optimization but don't directly tackle the core issue of exceeding the context window.
9 / 10
David (in a PR description): 'To improve summarization accuracy, we're employing context compression techniques. This involves distilling key information from the original document to reduce redundancy and focus on the most relevant details before feeding it into the model.' What is the primary goal of 'context compression' in this scenario?
Context compression aims to *reduce* the size of the input data – specifically, the content within the context window – by extracting essential information. This is done to improve both processing efficiency and summary accuracy because a smaller, more focused context reduces noise and irrelevant details that can mislead the model. Options A, C, and D describe different techniques or outcomes.
10 / 10
A developer is investigating slow response times from a chatbot after updating its knowledge base. The system logs show the model processing increasingly long input sequences. What is the most likely root cause related to the context window?
'Context dilution' describes the phenomenon where a model's ability to understand and reason about information diminishes as the length of the context window increases. The longer the sequence, the greater the difficulty for the model to maintain consistent attention and track relationships between different parts of the input. Options A, B, and D represent alternative causes of slow response times.
What will I practice in "Context Window Vocabulary"?
This is a Prompt Engineering Language exercise set. It walks through 10 scenario-based multiple-choice questions built around real usage of prompt engineering language terminology that IT professionals encounter on the job.
Is this exercise free to use?
Yes. Every exercise on CoderSlingo, including this one, is free to complete with no account, sign-up, or paywall.
How many questions are in this exercise?
This set contains 10 questions. Each one shows immediate feedback and a detailed explanation after you answer, so you learn the correct usage right away rather than waiting for a final score.
Do I need prior experience to complete this exercise?
No prior experience is required. Each question includes a full explanation covering the reasoning behind the correct answer, so the exercise itself teaches the prompt engineering language vocabulary as you go.
Can I retry the exercise if I get questions wrong?
Yes — use the "Try again" button on the results screen to reset your answers and go through all the questions again. There is no limit on attempts.
Is my progress saved?
Your answers and score for the current session are tracked in the browser as you go. No account or login is needed, and there is nothing to install.
What if I don't understand a term used in a question?
Read the explanation shown after you answer each question — it breaks down the correct term in plain English with a real-world example. You can also check the site Glossary for quick definitions.
How is this different from reading a blog article on the topic?
Exercises like this one are interactive drills that test and reinforce specific vocabulary through multiple-choice questions, while blog articles explain concepts in prose. Practising here after reading builds active recall, not just passive recognition.
Where can I find more Prompt Engineering Language exercises?
See the Prompt Engineering Language exercises hub for the full set of related pages, or browse all exercise categories from the main Exercises index.
Can I use this exercise to prepare for a technical interview?
Yes — prompt engineering language vocabulary comes up often in technical discussions and interviews. Pair this exercise with our dedicated Interview Preparation section for role-specific practice.