Not yet counting in evidence decisions. This row remains available for assessment, but does not currently carry a counting evidence result.
What was measured
token cost · token_delta How does the wording change tokenizer units for the declared tokenizer population?
Reported result
-28.466666666667 tokens per declared item Reported interval: -29.366666666667 to -28.466666666667.
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 3 declared conditions
by-construction, by-rule, in-practice
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
Awaiting independent settlement
An original reports one result. It does not confirm itself.
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?
Awaiting independent settlement.
An original reports one result. It does not confirm itself.
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
by-construction
-39.9
Not recorded
by-rule
-23.4
Not recorded
in-practice
-22.1
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 30 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 · utf8-codec
English input
Messages emitted by Codec Alder are valid UTF-8 because of how the system is built: its unchanged implementation cannot emit any other encoding. While it remains unchanged, an exception cannot occur; a non-UTF-8 message would falsify this claim or prove the system changed.
Ainglish input
Messages emitted by Codec Alder are valid UTF-8 by-construction.
Condition
by-construction
Input 2 · fixed-width-id
English input
IDs from Generator Birch are 128 bits long because of how the system is built: its unchanged generator always emits exactly 128 bits. While it remains unchanged, an exception cannot occur; an ID of any other length would falsify this claim or prove the system changed.
Ainglish input
IDs from Generator Birch are 128 bits long by-construction.
Condition
by-construction
Input 3 · four-field-parser
English input
Records accepted by Parser Cedar have exactly four fields because of how the system is built: its unchanged grammar accepts only four-field records. While it remains unchanged, an exception cannot occur; an accepted record with another field count would falsify this claim or prove the system changed.
Ainglish input
Records accepted by Parser Cedar have exactly four fields by-construction.
Condition
by-construction
Input 4 · filter-route
English input
Packets leaving Gateway Dune pass through filter F9 because of how the system is built: its unchanged topology has no route around filter F9. While it remains unchanged, an exception cannot occur; a departing packet that bypassed F9 would falsify this claim or prove the system changed.
Ainglish input
Packets leaving Gateway Dune pass through filter F9 by-construction.
Condition
by-construction
Input 5 · integer-counter
English input
Counter Elm is integer-valued because of how the system is built: its unchanged representation has no fractional state. While it remains unchanged, an exception cannot occur; a fractional counter value would falsify this claim or prove the system changed.
Ainglish input
Counter Elm is integer-valued by-construction.
Condition
by-construction
Input 6 · one-way-pipe
English input
Pipe Flax is one-way because of how the system is built: its unchanged valve geometry cannot carry reverse flow. While it remains unchanged, an exception cannot occur; reverse flow through the pipe would falsify this claim or prove the system changed.
Ainglish input
Pipe Flax is one-way by-construction.
Condition
by-construction
Recorded input digest: eb2f00d9b8433d326c0142431542c166b53f2fcb258413eb1a6cc3e6772d3de6
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
registered regime marker versus its complete English meaning, including what an exception would establish and who would owe
Tested population
30 frozen complete standing-property claims, 10 per regime across 30 distinct new domains
Unit tested
one complete standing-property claim with exception semantics
How results combine
equal-pair mean per tokenizer over all 30 claims, then the least-favourable maximum tokenizer mean; retain all three equal-weight regimes separately
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.
