Counts in current evidence decisions. This row currently contributes to evidence decisions. Its direction is separate from whether the proposal is ready for adoption.
What was measured
token cost · token_delta How does the wording change tokenizer units for the declared tokenizer population?
Reported result
-36 tokens per declared item Reported interval: -37.5 to -36.
Fewer tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.
English comparison
English comparison not recorded as a structured label
Declared by the submitter; not a certification of equivalent information.
Tokenizer conditions
Literal encoding cost on the named current tokenizers; not comprehension.
Named instruments
cl100k_base, o200k_base, p50k_base
Reader population not separately declared.
Conditions covered
Separate outcomes retained for all 2 declared conditions
no-delegation, one-hop-delegation-allowed
An overall average can hide a weak condition. A condition list is not proof that every form or claim in the proposal was tested.
Settlement role
Confirmed by eligible settlement
Eligible fresh-input replications currently give this original a settlement majority.
Is this result within the cost allowance?
No numerical allowance is available in this proposal’s current structured evidence declaration. A prose prediction is not silently converted into a bound.
This compares Ainglish minus English with the current declaration, which may differ from the declaration when the result was filed. It checks the headline only: inspect any required per-form and per-tokenizer results too.
Has the original estimate been independently reproduced?
Confirmed by eligible settlement.
Eligible fresh-input replications currently give this original a settlement majority.
Reproduction asks whether fresh-input findings agree under the settlement rule. It does not ask whether either value satisfies the cost allowance.
Being within the cost allowance is not a completed prerequisite. Reproducing an original estimate is a separate check, not proof that the allowance is met. Current evidence status, settlement and every declared result still determine readiness.
How can one check pass while the other does not?
For example, an allowance of at most +3 tokens and an original estimate of +3 ask different questions. A replication of −0.5 is within that allowance but may disagree with the original. A replication of +3.25 may reproduce +3 within the settlement tolerance while exceeding the allowance.
These are illustrative numbers, not a new settlement rule. A cost saving is not a comprehension result, and a reproduced premium does not by itself mean a proposal should be adopted or rejected.
Does the overall result hide differences between conditions?
Every stored condition, without new pooling. Differences and intervals use tokens. Condition names come from the frozen experiment.
Condition
Reported difference
Reported interval
no-delegation
-36
Not recorded
one-hop-delegation-allowed
-36
Not recorded
A missing condition interval is not zero uncertainty. An overall interval cannot substitute for agreement in every load-bearing condition.
Inspect actual inputs and recorded answers
The comparison label is the submitter’s declaration, not a semantic certification. Check that both versions preserve the information needed to answer the same question.
Numbers count only readable inputs attached to this receipt. They are not the experiment’s declared sample size or the number of reader calls.
Showing 1–6 of 24 readable, inline study items, in stored order—not a selection of successes. 0 control items are kept separate.
Each result has its own input pages. Positions across the two studies do not imply matched cases.
Input 1 · soil-custody-direct
English input
The responsible principal is required to certify the soil-sample chain of custody for lot Amber. The responsible principal must perform every completion-bearing part without assigning one to another principal; controlled deterministic tools and advice remain allowed when no other principal is made responsible for part of the work.
Ainglish input
req: certify the soil-sample chain of custody for lot Amber, no-delegation.
Condition
no-delegation
Input 2 · soil-custody-one-hop
English input
The responsible principal is required to certify the soil-sample chain of custody for lot Amber. The responsible principal may, but need not, assign work to any number of immediate delegates; those delegates may not delegate further, the original principal remains accountable for integration and completion, and no underlying authority is expanded.
Ainglish input
req: certify the soil-sample chain of custody for lot Amber, one-hop-delegation-allowed.
Condition
one-hop-delegation-allowed
Input 3 · pump-logs-direct
English input
The responsible principal is required to reconcile the dialysis-pump maintenance logs for ward Birch. The responsible principal must perform every completion-bearing part without assigning one to another principal; controlled deterministic tools and advice remain allowed when no other principal is made responsible for part of the work.
Ainglish input
req: reconcile the dialysis-pump maintenance logs for ward Birch, no-delegation.
Condition
no-delegation
Input 4 · pump-logs-one-hop
English input
The responsible principal is required to reconcile the dialysis-pump maintenance logs for ward Birch. The responsible principal may, but need not, assign work to any number of immediate delegates; those delegates may not delegate further, the original principal remains accountable for integration and completion, and no underlying authority is expanded.
Ainglish input
req: reconcile the dialysis-pump maintenance logs for ward Birch, one-hop-delegation-allowed.
Condition
one-hop-delegation-allowed
Input 5 · refund-trace-direct
English input
The responsible principal is required to trace refund anomaly R-418 through the signed settlement records. The responsible principal must perform every completion-bearing part without assigning one to another principal; controlled deterministic tools and advice remain allowed when no other principal is made responsible for part of the work.
Ainglish input
req: trace refund anomaly R-418 through the signed settlement records, no-delegation.
Condition
no-delegation
Input 6 · refund-trace-one-hop
English input
The responsible principal is required to trace refund anomaly R-418 through the signed settlement records. The responsible principal may, but need not, assign work to any number of immediate delegates; those delegates may not delegate further, the original principal remains accountable for integration and completion, and no underlying authority is expanded.
Ainglish input
req: trace refund anomaly R-418 through the signed settlement records, one-hop-delegation-allowed.
Condition
one-hop-delegation-allowed
Recorded input digest: b76115073853922ec75fa68d375f700b39be8d26d3c701a5c4a59fc6d1b4fb60
Prompts, reference material and other context can live elsewhere in the specification. Inputs and keys alone do not reconstruct every reader call or establish a fair comparison.
Declared population, method and retained outcomes
Compared with
no-delegation / one-hop-delegation-allowed versus complete careful English preserving responsibility, redelegation and authority limits
Tested population
24 frozen complete assignments: both delegation states crossed over 12 new operational contexts in 12 domains
Unit tested
one complete assignment with its delegation-depth condition
How results combine
equal item mean per tokenizer, then least-favourable maximum tokenizer mean; retain equal-weight literal-form strata
These are the study author’s declarations. A finding applies to this tested scope; this summary does not establish that another study is comparable.
Absolute arm results, reader-specific results and condition results below are retained values, not a newly pooled analysis. Accuracy arms use fractions from 0 to 1; their difference uses percentage points.
Different wording, readers, exposure or populations can legitimately produce different results. A visible reference is not training the model’s weights. Current models and tokenizers have learned English; future Ainglish-trained performance remains a research question.
