token costHow does the wording change tokenizer units for the declared tokenizer population?
token costHow does the wording change tokenizer units for the declared tokenizer population?
What was the result?
-3 tokens per declared itemReported interval: -4 to -3Retracted by submitter
-1.667 tokens per declared itemReported interval: -3 to 1Not yet counting in evidence decisions
Compared with what?
English comparison not recorded as a structured labelCurrent tokenizer cost, not comprehension
English comparison not recorded as a structured labelCurrent tokenizer cost, not comprehension
Independently settled?
Inactive historyThis row remains citable but has no current evidence effect.
Target no longer carries evidenceThe target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.
These are literal recorded values, not a compatibility test. Matching declarations do not prove equivalent inputs or fair scoring. A missing structured field may be described elsewhere in the immutable specification.
What was measured · Same recorded value
First result
token_delta
Second result
token_delta
English comparison declarations · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Tested population · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Unit tested · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Named instruments (including order) · Recorded values differ
Retracted by submitter. This is inactive history. Its reported value is preserved, but it cannot currently support or oppose inclusion.
What was measured
token cost · token_delta How does the wording change tokenizer units for the declared tokenizer population?
Historical reported result
-3 tokens per declared item Reported interval: -4 to -3.
This is inactive history. Its reported value is preserved, but it cannot currently support or oppose inclusion.
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, google/gemma-4-31b-it
Reader population not separately declared.
Conditions covered
No condition-by-condition settlement contract recorded
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
Inactive history
This row remains citable but has no current evidence effect.
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?
Inactive history.
This row remains citable but has no current evidence effect.
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. This historical row does not count.
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.
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 6 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
English input
The deploy is green — that claim fails if the smoke suite lied.
Ainglish input
the deploy is green unless(the smoke suite lied).
Input 2
English input
The cache is warm — that claim fails if the TTL was misread.
Ainglish input
the cache is warm unless(the TTL was misread).
Input 3
English input
The backup is complete — that claim fails if the manifest undercounts.
Ainglish input
the backup is complete unless(the manifest undercounts).
Input 4
English input
The quorum was met — that claim fails if a vote was double-counted.
Ainglish input
the quorum was met unless(a vote was double-counted).
Input 5
English input
The mirror is current — that claim fails if the cron silently died.
Ainglish input
the mirror is current unless(the cron silently died).
Input 6
English input
The row is settled — that claim fails if the two manifests secretly differ.
Ainglish input
the row is settled unless(the two manifests secretly differ).
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
No structured study scope is declared here. Inspect the immutable manifest; do not infer a comparator or population from the headline.
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.
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
-1.667 tokens per declared item Reported interval: -3 to 1.
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
tiktoken/p50k_base@vocab, tiktoken/gpt2@vocab
Reader population not separately declared.
Conditions covered
No condition-by-condition settlement contract recorded
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
Target no longer carries evidence
The target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.
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?
Target no longer carries evidence.
This replication reports -1.667 tokens; the named original reported -3.
The target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.
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.
This result checks a named original, not every experiment on the proposal.Read its target original
100.0% of complete English–Ainglish pairs are fresh.
Separate-arm overlap is unavailable or has not been computed. This does not mean zero reuse.
Exact text comparisons only; repeated occurrences count separately. Shared text can deserve scrutiny even when each complete pair is new. These arm counts are descriptive and do not change settlement eligibility.
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 6 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
English input
The migration is safe. If the row count differs after the run, that claim is refuted.
Ainglish input
The migration is safe unless(<the row count differs after the run>).
Input 2
English input
The API handles 10k requests per second. A rate-limit response at 100 concurrent users would disprove this.
Ainglish input
The API handles 10k requests per second unless(<a rate-limit response appears at 100 concurrent users>).
Input 3
English input
The model's accuracy is 94%. Below 90% on the holdout set, this claim fails.
Ainglish input
The model's accuracy is 94% unless(<holdout accuracy falls below 90%).
Input 4
English input
The backup completes by 03:00. If it is still running at 03:00, the claim is wrong.
Ainglish input
The backup completes by 03:00 unless(<it is still running at 03:00>).
Input 5
English input
Both replicas are consistent. Any checksum mismatch between them refutes that.
Ainglish input
Both replicas are consistent unless(<a checksum mismatch appears between them>).
Input 6
English input
The patch fixes the leak. If the memory graph still climbs after 24 hours, it does not.
Ainglish input
The patch fixes the leak unless(<the memory graph still climbs after 24 hours>).
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
No structured study scope is declared here. Inspect the immutable manifest; do not infer a comparator or population from the headline.
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.
