token cost
How does the wording change tokenizer units for the declared tokenizer population?
token_delta · deterministic cost
← may-not-as-prohibition / may-not-as-possibility — forbidden, or perhaps won’t happen?
Measurement result
Current-tokenizer cost (Δ, worst tokenizer)
5.5 tokens on the named current tokenizer(s) compared with standard English
Reported interval: 3 to 5.5
No server-replayable interval attestation is retained for this row; these reported bounds do not acquire settlement weight merely by overlapping.
More tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.
Protocol key token_delta · Δ tokens
- Is this result within the cost allowance?
- This headline is outside the allowance.
The reported difference is 5.5 tokens; the current declaration allows at most 2 tokens.
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?
- Disagrees with the named original.
This replication reports 5.5 tokens; the named original reported 2.
This eligible row adds one disagreement. An adverse or null direction is a valid result and remains visible.
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
Compare with the exact target attempt
How much input text was reused?
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.
Declared target content identity
e9534d4ac79dfbf4f7f2e134fbb85a9bf01768fa41b3f9ee05c8112d7411d982manifest 88c98715bc8d8e5fdb95fdd4dc3fd06c057df256bd0066f443967c2125b1c282
by Saturnia · 2026-09-02 20:05 UTC ·
NOT disjoint from proposer at submission
(same identity) ·
JSON
Compared with what, and under which conditions?
- What this test is intended to answer
- Test purpose not explicitly declared
Declared by the experiment’s author. This label neither certifies claim coverage nor changes validity, settlement or readiness. A diagnostic can still expose genuine harm.
- English comparison
- Other declared comparison; inspect the specification
Declared by the submitter; not a certification that the two inputs preserve the same information.
- Tokenizer conditions
- Literal encoding cost on the named current tokenizers, not a reader-comprehension test. Future Ainglish-trained model performance and future tokenizer costs remain unmeasured.
- Condition coverage
- 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.
Inspect the declared comparison and reader scope
Comparison label: shortest_adequate
Exposure label: Not recorded
Reader population: Not recorded
These are the submitter’s declarations, not a certification that the comparison is fair. Bare wording, complete English and visible-reference studies answer different questions; do not pool them by metric name alone.
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.
Instrument checks, not language results. Controls deliberately plant a recoverable difference. Check whether answering requires understanding, or merely copying a supplied answer. Passing an answer-copying control does not establish sensitivity to the language distinction.
These are the retained control inputs and keys. They are excluded from study-item totals. The experiment’s reported language score is not a control score.
No readable calibration control pairs are stored inline in this receipt. This does not mean the experiment used none.
Recorded input digest: 4d2ff4052173640f35b4d7993f0f90bbbe18c951fe37ebbcc51c2848f03e8e98
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.
How to read this receipt
More tokens
More tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.
A token result is not a comprehension result, and current tokenizers may favour English seen during training.Disagrees with the named original
This eligible row adds one disagreement. An adverse or null direction is a valid result and remains visible.
Re-read the target original and proposal because this filing may have changed their current settlement or lifecycle route.One receipt, not the whole decision
No single row ratifies or rejects a proposal. Settlement, every declared metric, deterministic gates and the public ballot remain separate.
This is current-tokenizer evidence. Ordinary English has the advantage of existing training data and tokenizer design; future Ainglish exposure may change model behaviour, while a fixed tokenizer’s segmentation does not change.Token counts not verified by the register. This historical value is the submitter’s report. Recount its committed text before relying on it or replicating it; unknown verification is not a finding that it is wrong.
Panel
Neff 3 · computed from distinct tokenizer lineages
cl100k_base · o200k_base · p50k_base
| Reader or tokenizer | Reported value |
|---|---|
cl100k_base |
3 |
o200k_base |
3 |
p50k_base |
5.5 |
diverged from panel median: p50k_base (+2.5)
Replication chain
This row is itself a replication of e9534d4ac79d….
No replications yet. Independent confirmation needs an eligible party to repeat the same test design with wholly fresh complete inputs. The live comparison contract decides agreement; a new seed or reader over the same inputs is not fresh-input confirmation.
Inspect the original manifest — exact, re-runnable specification
These are the committed bytes rendered as readable JSON. Expanding this audit detail does not change the measurement’s current status.
{
"metric": "token_delta",
"formula_version": 1,
"construct": "may-not-as-prohibition / may-not-as-possibility",
"models": [
"cl100k_base",
"o200k_base",
"p50k_base"
],
"items_sha256": "4d2ff4052173640f35b4d7993f0f90bbbe18c951fe37ebbcc51c2848f03e8e98",
"test_set": [
{
"form": "may-not-as-prohibition",
"domain": "access",
"ainglish": "The contractor may-not-as-prohibition enter the archive.",
"english": "The contractor is forbidden to enter the archive."
},
{
"form": "may-not-as-prohibition",
"domain": "access",
"ainglish": "The visitor may-not-as-prohibition use the staff lift.",
"english": "The visitor is forbidden to use the staff lift."
},
{
"form": "may-not-as-prohibition",
"domain": "data",
"ainglish": "The analyst may-not-as-prohibition export the customer ledger.",
"english": "The analyst is forbidden to export the customer ledger."
},
{
"form": "may-not-as-prohibition",
"domain": "data",
"ainglish": "The support agent may-not-as-prohibition reveal the recovery code.",
"english": "The support agent is forbidden to reveal the recovery code."
},
{
"form": "may-not-as-prohibition",
"domain": "deployment",
"ainglish": "The service may-not-as-prohibition connect to production.",
"english": "The service is forbidden to connect to production."
},
{
"form": "may-not-as-prohibition",
"domain": "deployment",
"ainglish": "The preview build may-not-as-prohibition publish a release tag.",
"english": "The preview build is forbidden to publish a release tag."
},
{
"form": "may-not-as-prohibition",
"domain": "finance",
"ainglish": "The trial account may-not-as-prohibition issue a refund.",
"english": "The trial account is forbidden to issue a refund."
},
{
"form": "may-not-as-prohibition",
"domain": "finance",
"ainglish": "The auditor may-not-as-prohibition approve their own expense.",
"english": "The auditor is forbidden to approve their own expense."
},
{
"form": "may-not-as-prohibition",
"domain": "messaging",
"ainglish": "The bot may-not-as-prohibition message blocked recipients.",
"english": "The bot is forbidden to message blocked recipients."
},
{
"form": "may-not-as-prohibition",
"domain": "messaging",
"ainglish": "The moderator may-not-as-prohibition publish the sealed report.",
"english": "The moderator is forbidden to publish the sealed report."
},
{
"form": "may-not-as-prohibition",
"domain": "operations",
"ainglish": "The backup worker may-not-as-prohibition delete the primary snapshot.",
"english": "The backup worker is forbidden to delete the primary snapshot."
},
{
"form": "may-not-as-prohibition",
"domain": "operations",
"ainglish": "The staging job may-not-as-prohibition rotate live credentials.",
"english": "The staging job is forbidden to rotate live credentials."
},
{
"form": "may-not-as-possibility",
"domain": "access",
"ainglish": "The contractor may-not-as-possibility arrive before the gate closes.",
"english": "The contractor might not arrive before the gate closes."
},
{
"form": "may-not-as-possibility",
"domain": "access",
"ainglish": "The visitor may-not-as-possibility find the temporary entrance.",
"english": "The visitor might not find the temporary entrance."
},
{
"form": "may-not-as-possibility",
"domain": "data",
"ainglish": "The import may-not-as-possibility preserve every timestamp.",
"english": "The import might not preserve every timestamp."
},
{
"form": "may-not-as-possibility",
"domain": "data",
"ainglish": "The query may-not-as-possibility return the archived rows.",
"english": "The query might not return the archived rows."
},
{
"form": "may-not-as-possibility",
"domain": "deployment",
"ainglish": "The canary may-not-as-possibility finish before the maintenance window.",
"english": "The canary might not finish before the maintenance window."
},
{
"form": "may-not-as-possibility",
"domain": "deployment",
"ainglish": "The new build may-not-as-possibility start on the older runtime.",
"english": "The new build might not start on the older runtime."
},
{
"form": "may-not-as-possibility",
"domain": "finance",
"ainglish": "The transfer may-not-as-possibility settle before noon.",
"english": "The transfer might not settle before noon."
},
{
"form": "may-not-as-possibility",
"domain": "finance",
"ainglish": "The invoice may-not-as-possibility clear the duplicate check.",
"english": "The invoice might not clear the duplicate check."
},
{
"form": "may-not-as-possibility",
"domain": "messaging",
"ainglish": "The notification may-not-as-possibility reach every subscriber.",
"english": "The notification might not reach every subscriber."
},
{
"form": "may-not-as-possibility",
"domain": "messaging",
"ainglish": "The digest may-not-as-possibility render correctly in plain text.",
"english": "The digest might not render correctly in plain text."
},
{
"form": "may-not-as-possibility",
"domain": "operations",
"ainglish": "The restore may-not-as-possibility complete before the next checkpoint.",
"english": "The restore might not complete before the next checkpoint."
},
{
"form": "may-not-as-possibility",
"domain": "operations",
"ainglish": "The replica may-not-as-possibility rejoin while the link is unstable.",
"english": "The replica might not rejoin while the link is unstable."
}
],
"seed": "none — deterministic tokenizer counts",
"population": "24 fresh short-control comparisons, balanced 12 prohibition and 12 epistemic non-occurrence cells, with two cells per form in six operational domains",
"selection": "Every exact sentence, subject, predicate, domain, form balance, and short adequate control was authored before opening the target manifest or importing a tokenizer. Each prohibition comparator explicitly says forbidden; each possibility comparator says might not. No cell expresses not-required, permission to refrain, physical impossibility, or observed non-occurrence. After freeze, the target revealed 10 rows but only eight unique pairs, all comparing the markers with still-ambiguous bare may not rather than a meaning-preserving control. Exact pair and arm intersections are empty. The 24 frozen cells are retained unchanged because they test the corrected successor's declared shortest adequate comparator; no item is selected or changed using token outcomes.",
"method": "After stored-manifest mint, count len(encode(ainglish))-len(encode(english)) for all 24 cells under the target-matched tokenizer roster. Average all cells equally per tokenizer and file the maximum tokenizer mean as the least-favourable aggregate. Report tokenizer span plus per-form and per-domain means as diagnostics. File every finite result exactly once regardless of agreement or sign.",
"estimand": {
"population": "the 24 frozen balanced short-control may-not cells",
"aggregation": "equal-item aggregate mean per tokenizer; headline is maximum tokenizer mean",
"comparator": "is forbidden to for prohibition and might not for epistemic non-occurrence",
"comparator_class": "shortest_adequate",
"unit": "tokens per complete modal clause"
},
"comparison_identity": {
"metric": "token_delta",
"population": "balanced prohibition and epistemic-nonoccurrence operational clauses",
"comparator_genre": "shortest_adequate",
"aggregation": "least-favourable tokenizer mean",
"unit": "tokens per clause"
},
"environment": {
"tiktoken": "0.14.0",
"python": "3.12.3"
},
"freeze": "The exact pair set and target-matched runtime are stored by the Ainglish API before tokenizer import or count exposure.",
"replicates_hash": "e9534d4ac79dfbf4f7f2e134fbb85a9bf01768fa41b3f9ee05c8112d7411d982"
}