Retrieval and embeddings · Lesson 11 of 16 · 6 min

Inspect duplicate chunks

Duplicate chunks can crowd a result list and reduce source diversity. Review stable document/chunk identifiers before deciding whether two similar passages are redundant.

Open the lab Course map

Mastery 0/3Run itExplain itApply it

learner@clairos:~ $

↑ history · Tab complete · Ctrl+C cancel · Esc, Tab to leaveExit 0

Your objectives

Hints are always available. Wrong turns never cost XP.

1
Read the case evidence
Read /opt/ai-lab/cases/aiops-retrieval-embeddings-011.json. Separate the recorded facts from your assumptions.
Hint & worked example
Duplicate chunks can crowd a result list and reduce source diversity. Review stable document/chunk identifiers before deciding whether two similar passages are redundant.
The complete case JSON is read successfully.
```
cat /opt/ai-lab/cases/aiops-retrieval-embeddings-011.json
```
Common mistake to avoid: Five ranked rows always represent five independent sources.
2
Isolate the deciding signal
Select the duplicate_count field from the case JSON. What does it establish—and what does it not establish?
Hint & worked example
Review deduplication so repeated chunks do not consume the whole evidence list.
A successful JSON query isolates duplicate_count.
```
jq .duplicate_count /opt/ai-lab/cases/aiops-retrieval-embeddings-011.json
```
Common mistake to avoid: Using a different case or an unrelated signal as the justification.
3
Record your operational decision
Choose the justified code: deduplicate-results, change-without-evidence, or ignore-the-signal. Save that one code followed by a newline in /home/learner/lab/aiops-retrieval-embeddings-011-decision.txt.
Hint & worked example
Review deduplication so repeated chunks do not consume the whole evidence list.
The decision artifact contains the correct code, not just a matching command in history.
```
printf '%s\n' 'deduplicate-results' > /home/learner/lab/aiops-retrieval-embeddings-011-decision.txt
```
Common mistake to avoid: Five ranked rows always represent five independent sources.

Understand the pattern

Duplicate chunks can crowd a result list and reduce source diversity. Review stable document/chunk identifiers before deciding whether two similar passages are redundant.

Interpret the evidence

Review deduplication so repeated chunks do not consume the whole evidence list.

A tempting mistake

Five ranked rows always represent five independent sources.

All names, measurements, prices, versions, and thresholds in this exercise are fictional training data. The evidence supports this bounded scenario, not a claim about a live provider.

Worked example and expected result

$ cat /opt/ai-lab/cases/aiops-retrieval-embeddings-011.json

{
  "fixture": "Fictional training data; not a live measurement.",
  "results": 5,
  "unique_chunk_ids": 2,
  "duplicate_count": 3
}

Exit status: 0. The command completed successfully in the training fixture.

$ jq .duplicate_count /opt/ai-lab/cases/aiops-retrieval-embeddings-011.json

Exit status: 0. The command completed successfully in the training fixture.

$ printf '%s\n' 'deduplicate-results' > /home/learner/lab/aiops-retrieval-embeddings-011-decision.txt

(no output; exit 0)

Exit status: 0. The command completed successfully in the training fixture.

Try a variation

Name the assumption that would have to change for “deduplicate-results” to stop being the justified action. Do not silently replace missing evidence with a guess.

Explain it

What does the evidence in “Inspect duplicate chunks” justify?Make the change now and gather the missing evidence afterward.Review deduplication so repeated chunks do not consume the whole evidence list.Five ranked rows always represent five independent sources.

Apply it elsewhere

A colleague concludes: “Five ranked rows always represent five independent sources.” What should you do?Reject that conclusion; return to the case evidence and its stated limits.Accept it because the training command completed.Treat this fixture as a live measurement.

Keep the momentum.

Complete the lab, then explain and apply the pattern.

Next: Review metadata filters Try an incident drill

Anonymous lesson and mastery progress lasts for this session. Sign in to keep synced lesson XP between sessions; mastery stays device-local. No speed bonus, lost lives, or hint penalties.

Keep learning

Case methodology: evidence, limits, decision