deployment mismatch concept
| Definition | The failure in which the consumer that was evaluated is not the consumer that was deployed, or time moved between the two. Chapters 1 and 13. Also deployment mean, the consumer that was deployed. |
|---|---|
| Example | The harness read a direction the deployed consumer never sees, and its 0.909 became 0.736 in deployment. |
| Book | Data Mining as Observation, draft 0.2, commit f3914f0; entry id deployment-mismatch, kind concept. |
| Status | no ledger row names this entry. Corrections: none recorded. |
| Defining equation | none |
| Assumptions and scope |
|
| Prior art | none recorded |
| Evidence | observation-theory-campaigns/experiments/LLM-EVAL-TRACK.md:1-50, observation-theory-campaigns/analysis/llm/XPROTO-LLM-graded.json, observation-theory-campaigns/experiments/SEALS.md:85, lean/DataMiningAsObservation/DeploymentMismatch.lean |
| Reviewed | not yet reviewed; generated 2026-09-10 from records at the commits on the provenance page. |
Equation
none
Conditions
- The failure in which the consumer that was evaluated is not the consumer that was deployed, or time moved between the two. A deployment score is a weighted mean over the slices actually served, so it lies between the worst slice and the best, and a benchmark drawn from the best slice overstates it unless every weighted slice matches.
- The chapter’s case scores 0.909 on the benchmark against a target of 0.8 and 0.736 on the deployment mean, and the six bars that separate the two are preregistered.
Conditions are curated in entries.toml rather than read
from a record.
Ledger
none
First stated
Chapter 1 section 1.7 and chapter 13 of Data Mining as
Observation, with the program’s case in the radio slice sweep,
observation-theory-campaigns/analysis/llm/PREREG-XPROTO-LLM.md,
and the serving-stack measurement of ledger row GO-2/GO-12/GO-13
operational.
Measurements
| Where the book states it | Numbers, as the book’s sources table records them | Source |
|---|---|---|
| chapter 12 section 12.6 | XPROTO-LLM, benchmark 0.909, 0.920, 0.909, thirty slices, target 0.8, naive 0.333, aware 0.033, spread 0.380, deployment mean 0.736, six bars on three seeds, sealed 2026-08-25 at b61f7f1 | observation-theory-campaigns/experiments/LLM-EVAL-TRACK.md:1-50;
observation-theory-campaigns/analysis/llm/XPROTO-LLM-graded.json;
observation-theory-campaigns/experiments/SEALS.md:85 |
Failures and corrections
none
Invariance envelope
none declared
Machine checked
lean/DataMiningAsObservation/DeploymentMismatch.lean,
theorems deployment_le_max, min_le_deployment,
deployment_lt_max_of_gap, book_numbers, at
observation-data-mining f3914f0; what the check covers is stated in the
book’s appendix
C.
Used in
Data Mining as Observation chapters 1, 12.
Related
observer; certificate; coherence time; min-over-strata.
See also
Book equations stated beside the entry’s terms, not defining it: 13.1.
Ledger rows that cite the entry’s records without naming it: OT-4, GO-2/GO-12/GO-13 operational (KV serving, 077).
Sources-table rows that share a record with the entry without naming it: chapter 13 section 13.6.
Status
Generated 2026-09-10 by encyclopedia/generate.py; book
at observation-data-mining f3914f0; the commit of every record is listed
in the encyclopedia’s provenance.