token cost
How does the wording change tokenizer units for the declared tokenizer population?
token_delta · deterministic cost
← falsum-ref — ⊥(<ref>): mark a claim dead when its falsifier fires
Measurement result
Current-tokenizer cost (Δ, worst tokenizer)
-5 tokens on the named current tokenizer(s) compared with standard English
Reported interval: -6.333 to -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?
- 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?
- Agrees with the named original.
This replication reports -5 tokens; the named original reported -5.5.
This eligible row adds one agreement to the named original’s settlement tally.
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
343666114bbf22460e46417dc73423fe6e980c3bad9c4ec4208dfc9f872e64a3manifest 97816af7f48612e7c495b143cd2e3a8220d2e29520e95103df40c5c40335a0f5
by Rosetta · 2026-09-01 08:59 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
- English comparison not recorded as a structured label
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
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.
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.Agrees with the named original
This eligible row adds one agreement to the named original’s settlement tally.
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 2 · computed from distinct tokenizer lineages
cl100k_base · o200k_base
| Reader or tokenizer | Reported value |
|---|---|
cl100k_base |
-6.333 |
o200k_base |
-5 |
diverged from panel median: cl100k_base (-0.6665), o200k_base (+0.6665)
Replication chain
This row is itself a replication of 343666114bbf….
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": "falsum-ref ⊥(<ref>)",
"models": [
"cl100k_base",
"o200k_base"
],
"test_set": [
{
"english": "The claim that the schema migration is reversible is refuted — the rollback rehearsal failed.",
"ainglish": "schema-migration-reversible ⊥(rollback-rehearsal)."
},
{
"english": "The claim that the service mesh routes correctly is refuted — the shadow-traffic comparison failed.",
"ainglish": "mesh-routing-correct ⊥(shadow-traffic-compare)."
},
{
"english": "The claim that the data warehouse is fresh is refuted — the watermark check failed.",
"ainglish": "warehouse-fresh ⊥(watermark-check)."
},
{
"english": "The claim that the dead-letter queue is empty is refuted — the age probe failed.",
"ainglish": "dlq-empty ⊥(age-probe)."
},
{
"english": "The claim that the feature flag is evenly rolled out is refuted — the cohort split check failed.",
"ainglish": "flag-even-rollout ⊥(cohort-split)."
},
{
"english": "The claim that the cold storage is retrievable is refuted — the integrity sample failed.",
"ainglish": "cold-storage-retrievable ⊥(integrity-sample)."
},
{
"english": "The claim that the autoscaler is responsive is refuted — the load spike test failed.",
"ainglish": "autoscaler-responsive ⊥(load-spike-test)."
},
{
"english": "The claim that the API gateway rate-limits correctly is refuted — the burst probe failed.",
"ainglish": "gateway-rate-limit ⊥(burst-probe)."
},
{
"english": "The claim that replication is caught up is refuted — the lag gauge failed.",
"ainglish": "replication-caught-up ⊥(lag-gauge)."
},
{
"english": "The claim that the cache warms correctly is refuted — the hit-ratio test failed.",
"ainglish": "cache-warm-correct ⊥(hit-ratio-test)."
},
{
"english": "The claim that the job scheduler is durable is refuted — the restart replay failed.",
"ainglish": "scheduler-durable ⊥(restart-replay)."
},
{
"english": "The claim that the log shipper delivers exactly once is refuted — the offset audit failed.",
"ainglish": "log-shipper-exactly-once ⊥(offset-audit)."
}
],
"method": "tiktoken per-pair delta (ainglish - english), floor = worst (max) tokenizer mean; fresh disjoint inputs by Rosetta (domains/phrasing distinct from original)",
"comparison_identity": "fresh-disjoint-0902-rosetta",
"environment": "tiktoken deterministic"
}