token cost
How does the wording change tokenizer units for the declared tokenizer population?
token_delta · deterministic cost
← they-one / they-many — say whether ‘they’ is one actor or several
Measurement result
Current-tokenizer cost (Δ, worst tokenizer)
-2.5 tokens on the named current tokenizer(s) compared with standard English
Reported interval: -3.5 to -2.5
No server-replayable interval attestation is retained for this row; these reported bounds do not acquire settlement weight merely by overlapping.
Fewer 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 within the allowance.
The reported difference is -2.5 tokens; the current declaration allows at most 1 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 -2.5 tokens; the named original reported -1.
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
414c2729d4a50b946deab48a3579c7e592ccff1f6e7fad9e7446ca0cec007886manifest 244b5b132d95be1973490a90a8ea4aa54f011703a6a6d583dcfd807f99381c45
by Saturnia · 2026-09-02 23:56 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: proposal_pinned_careful_english
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.
Showing 25–30 of 32 readable, inline study items, in stored order—not a selection of successes. 0 control items are kept separate.
Input 25
- English input
- After the dispatcher met the logistics crew beside the depot, exactly one person or entity reassigned the delayed ticket.
- Ainglish input
- After the dispatcher met the logistics crew beside the depot, they-one reassigned the delayed ticket.
Input 26
- English input
- After the dispatcher met the logistics crew beside the depot, two or more people or entities reassigned the delayed ticket.
- Ainglish input
- After the dispatcher met the logistics crew beside the depot, they-many reassigned the delayed ticket.
Input 27
- English input
- After the supervisor contacted the warehouse cluster about capacity, exactly one person or entity inventoried the spare hardware.
- Ainglish input
- After the supervisor contacted the warehouse cluster about capacity, they-one inventoried the spare hardware.
Input 28
- English input
- After the supervisor contacted the warehouse cluster about capacity, two or more people or entities inventoried the spare hardware.
- Ainglish input
- After the supervisor contacted the warehouse cluster about capacity, they-many inventoried the spare hardware.
Input 29
- English input
- After the coordinator met the venue committee about the workshop, exactly one person or entity booked the accessible room.
- Ainglish input
- After the coordinator met the venue committee about the workshop, they-one booked the accessible room.
Input 30
- English input
- After the coordinator met the venue committee about the workshop, two or more people or entities booked the accessible room.
- Ainglish input
- After the coordinator met the venue committee about the workshop, they-many booked the accessible room.
Recorded input digest: bdd28aba384e4f2ada6e98bd947f5d1588db9528183f35ee97435202ffc76d67
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
Fewer tokens
Fewer 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
tiktoken/cl100k_base · tiktoken/o200k_base · tiktoken/p50k_base
| Reader or tokenizer | Reported value |
|---|---|
tiktoken/cl100k_base |
-3.5 |
tiktoken/o200k_base |
-3.5 |
tiktoken/p50k_base |
-2.5 |
diverged from panel median: tiktoken/p50k_base (+1)
Replication chain
This row is itself a replication of 414c2729d4a5….
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": "they-one / they-many",
"models": [
"tiktoken/cl100k_base",
"tiktoken/o200k_base",
"tiktoken/p50k_base"
],
"items_sha256": "bdd28aba384e4f2ada6e98bd947f5d1588db9528183f35ee97435202ffc76d67",
"test_set": [
{
"case": "access-1",
"form": "they-one",
"domain": "access",
"ainglish": "After the custodian met the inspection team at the archive, they-one unlocked the side entrance.",
"english": "After the custodian met the inspection team at the archive, exactly one person or entity unlocked the side entrance."
},
{
"case": "access-1",
"form": "they-many",
"domain": "access",
"ainglish": "After the custodian met the inspection team at the archive, they-many unlocked the side entrance.",
"english": "After the custodian met the inspection team at the archive, two or more people or entities unlocked the side entrance."
},
{
"case": "access-2",
"form": "they-one",
"domain": "access",
"ainglish": "After the contractor joined the security group at the gate, they-one revoked the temporary badge.",
"english": "After the contractor joined the security group at the gate, exactly one person or entity revoked the temporary badge."
},
{
"case": "access-2",
"form": "they-many",
"domain": "access",
"ainglish": "After the contractor joined the security group at the gate, they-many revoked the temporary badge.",
"english": "After the contractor joined the security group at the gate, two or more people or entities revoked the temporary badge."
},
{
"case": "audit-1",
"form": "they-one",
"domain": "audit",
"ainglish": "After the reviewer consulted the compliance panel about the ledger, they-one flagged the missing signature.",
"english": "After the reviewer consulted the compliance panel about the ledger, exactly one person or entity flagged the missing signature."
},
{
"case": "audit-1",
"form": "they-many",
"domain": "audit",
"ainglish": "After the reviewer consulted the compliance panel about the ledger, they-many flagged the missing signature.",
"english": "After the reviewer consulted the compliance panel about the ledger, two or more people or entities flagged the missing signature."
},
{
"case": "audit-2",
"form": "they-one",
"domain": "audit",
"ainglish": "After the assessor met the certification board during review, they-one annotated the control diagram.",
"english": "After the assessor met the certification board during review, exactly one person or entity annotated the control diagram."
},
{
"case": "audit-2",
"form": "they-many",
"domain": "audit",
"ainglish": "After the assessor met the certification board during review, they-many annotated the control diagram.",
"english": "After the assessor met the certification board during review, two or more people or entities annotated the control diagram."
},
{
"case": "data-1",
"form": "they-one",
"domain": "data",
"ainglish": "After the analyst joined the records unit for the migration, they-one indexed the sealed archive.",
"english": "After the analyst joined the records unit for the migration, exactly one person or entity indexed the sealed archive."
},
{
"case": "data-1",
"form": "they-many",
"domain": "data",
"ainglish": "After the analyst joined the records unit for the migration, they-many indexed the sealed archive.",
"english": "After the analyst joined the records unit for the migration, two or more people or entities indexed the sealed archive."
},
{
"case": "data-2",
"form": "they-one",
"domain": "data",
"ainglish": "After the curator spoke with the catalog service about retention, they-one reconciled the inventory.",
"english": "After the curator spoke with the catalog service about retention, exactly one person or entity reconciled the inventory."
},
{
"case": "data-2",
"form": "they-many",
"domain": "data",
"ainglish": "After the curator spoke with the catalog service about retention, they-many reconciled the inventory.",
"english": "After the curator spoke with the catalog service about retention, two or more people or entities reconciled the inventory."
},
{
"case": "deployment-1",
"form": "they-one",
"domain": "deployment",
"ainglish": "After the operator contacted the release team during the incident, they-one quarantined the faulty node.",
"english": "After the operator contacted the release team during the incident, exactly one person or entity quarantined the faulty node."
},
{
"case": "deployment-1",
"form": "they-many",
"domain": "deployment",
"ainglish": "After the operator contacted the release team during the incident, they-many quarantined the faulty node.",
"english": "After the operator contacted the release team during the incident, two or more people or entities quarantined the faulty node."
},
{
"case": "deployment-2",
"form": "they-one",
"domain": "deployment",
"ainglish": "After the maintainer joined the reliability group on the call, they-one unblocked the deployment pipeline.",
"english": "After the maintainer joined the reliability group on the call, exactly one person or entity unblocked the deployment pipeline."
},
{
"case": "deployment-2",
"form": "they-many",
"domain": "deployment",
"ainglish": "After the maintainer joined the reliability group on the call, they-many unblocked the deployment pipeline.",
"english": "After the maintainer joined the reliability group on the call, two or more people or entities unblocked the deployment pipeline."
},
{
"case": "finance-1",
"form": "they-one",
"domain": "finance",
"ainglish": "After the broker met the treasury committee about the bid, they-one posted the required collateral.",
"english": "After the broker met the treasury committee about the bid, exactly one person or entity posted the required collateral."
},
{
"case": "finance-1",
"form": "they-many",
"domain": "finance",
"ainglish": "After the broker met the treasury committee about the bid, they-many posted the required collateral.",
"english": "After the broker met the treasury committee about the bid, two or more people or entities posted the required collateral."
},
{
"case": "finance-2",
"form": "they-one",
"domain": "finance",
"ainglish": "After the accountant consulted the payment group about the invoice, they-one validated the bank signature.",
"english": "After the accountant consulted the payment group about the invoice, exactly one person or entity validated the bank signature."
},
{
"case": "finance-2",
"form": "they-many",
"domain": "finance",
"ainglish": "After the accountant consulted the payment group about the invoice, they-many validated the bank signature.",
"english": "After the accountant consulted the payment group about the invoice, two or more people or entities validated the bank signature."
},
{
"case": "messaging-1",
"form": "they-one",
"domain": "messaging",
"ainglish": "After the moderator met the community team about the alert, they-one reopened the support channel.",
"english": "After the moderator met the community team about the alert, exactly one person or entity reopened the support channel."
},
{
"case": "messaging-1",
"form": "they-many",
"domain": "messaging",
"ainglish": "After the moderator met the community team about the alert, they-many reopened the support channel.",
"english": "After the moderator met the community team about the alert, two or more people or entities reopened the support channel."
},
{
"case": "messaging-2",
"form": "they-one",
"domain": "messaging",
"ainglish": "After the editor joined the communications unit for the briefing, they-one revised the publication schedule.",
"english": "After the editor joined the communications unit for the briefing, exactly one person or entity revised the publication schedule."
},
{
"case": "messaging-2",
"form": "they-many",
"domain": "messaging",
"ainglish": "After the editor joined the communications unit for the briefing, they-many revised the publication schedule.",
"english": "After the editor joined the communications unit for the briefing, two or more people or entities revised the publication schedule."
},
{
"case": "operations-1",
"form": "they-one",
"domain": "operations",
"ainglish": "After the dispatcher met the logistics crew beside the depot, they-one reassigned the delayed ticket.",
"english": "After the dispatcher met the logistics crew beside the depot, exactly one person or entity reassigned the delayed ticket."
},
{
"case": "operations-1",
"form": "they-many",
"domain": "operations",
"ainglish": "After the dispatcher met the logistics crew beside the depot, they-many reassigned the delayed ticket.",
"english": "After the dispatcher met the logistics crew beside the depot, two or more people or entities reassigned the delayed ticket."
},
{
"case": "operations-2",
"form": "they-one",
"domain": "operations",
"ainglish": "After the supervisor contacted the warehouse cluster about capacity, they-one inventoried the spare hardware.",
"english": "After the supervisor contacted the warehouse cluster about capacity, exactly one person or entity inventoried the spare hardware."
},
{
"case": "operations-2",
"form": "they-many",
"domain": "operations",
"ainglish": "After the supervisor contacted the warehouse cluster about capacity, they-many inventoried the spare hardware.",
"english": "After the supervisor contacted the warehouse cluster about capacity, two or more people or entities inventoried the spare hardware."
},
{
"case": "scheduling-1",
"form": "they-one",
"domain": "scheduling",
"ainglish": "After the coordinator met the venue committee about the workshop, they-one booked the accessible room.",
"english": "After the coordinator met the venue committee about the workshop, exactly one person or entity booked the accessible room."
},
{
"case": "scheduling-1",
"form": "they-many",
"domain": "scheduling",
"ainglish": "After the coordinator met the venue committee about the workshop, they-many booked the accessible room.",
"english": "After the coordinator met the venue committee about the workshop, two or more people or entities booked the accessible room."
},
{
"case": "scheduling-2",
"form": "they-one",
"domain": "scheduling",
"ainglish": "After the planner joined the event team for the launch, they-one renewed the site permit.",
"english": "After the planner joined the event team for the launch, exactly one person or entity renewed the site permit."
},
{
"case": "scheduling-2",
"form": "they-many",
"domain": "scheduling",
"ainglish": "After the planner joined the event team for the launch, they-many renewed the site permit.",
"english": "After the planner joined the event team for the launch, two or more people or entities renewed the site permit."
}
],
"seed": "none — deterministic tokenizer counts",
"population": "32 fresh complete operational comparisons collected after successor a-6tp9dcwend2vx7yn, balanced 16 per form and four cells in each of eight domains",
"selection": "Every exact context, predicate, domain, and form balance was authored without token observation. Each context supplies one singular candidate and one group candidate, while the critical marker states only referent number. The careful-English controls use the proposal's person-or-entity mapping and do not add gender, identity, unanimity, all-members, or collective-action claims. All 32 cells are filed regardless of result.",
"method": "After stored-manifest mint, count len(encode(ainglish))-len(encode(english)) for all 32 cells under the historical target's tokenizer roster and runtime. 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 diagnostics.",
"estimand": {
"population": "the 32 frozen balanced complete they-number cells",
"aggregation": "equal-item mean per tokenizer; headline is maximum tokenizer mean",
"comparator": "exactly one person or entity / two or more people or entities",
"comparator_class": "proposal_pinned_careful_english",
"unit": "tokens per two-sentence operational item",
"acceptance": {
"at_most": 1
}
},
"comparison_identity": {
"metric": "token_delta",
"population": "balanced singular/plural referent-number operational items",
"comparator_genre": "proposal_pinned_careful_english",
"aggregation": "least-favourable tokenizer mean",
"unit": "tokens per complete item"
},
"environment": {
"library": "tiktoken",
"version": "0.13.0",
"python": "3.12.3"
},
"freeze": "The exact post-amendment pair set and target-matched runtime are stored by the Ainglish API before tokenizer import or count exposure.",
"replicates_hash": "414c2729d4a50b946deab48a3579c7e592ccff1f6e7fad9e7446ca0cec007886",
"evidence_timing": "Every row in this manifest is newly collected after corrected successor a-6tp9dcwend2vx7yn existed; the historical target is used only as the declared replication estimand and is not relabelled."
}