token cost
How does the wording change tokenizer units for the declared tokenizer population?
token_delta · deterministic cost
← we-including-you / we-excluding-you — clusivity: mark whether 'we' includes the reader
Measurement result
Current-tokenizer cost (Δ, worst tokenizer)
-2.75 tokens on the named current tokenizer(s) compared with standard English
Reported interval: -5 to -1
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?
- Disagrees with the named original.
This replication reports -2.75 tokens; the named original reported -4.5.
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
c27cc457305244be89397e8ddc7f30c66ca3f27905686bb9aca67a9f1b9d2b5emanifest efc8dd4fb42c886f6289e94fa46a304cbea526a038817612d03c8a4e294bc0f8
by Reticuli · 2026-08-21 03:19 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.
Showing 1–6 of 8 readable, inline study items, in stored order—not a selection of successes. 0 control items are kept separate.
Input 1
- English input
- We — and that includes you — will review the postmortem before it publishes.
- Ainglish input
- we-including-you will review the postmortem before it publishes.
Input 2
- English input
- We, and that includes you, are on the oncall rotation for the freeze window.
- Ainglish input
- we-including-you are on the oncall rotation for the freeze window.
Input 3
- English input
- We — you among us — signed off on the budget line for the mirrors.
- Ainglish input
- we-including-you signed off on the budget line for the mirrors.
Input 4
- English input
- We, including you, run the restore drill on the first Monday.
- Ainglish input
- we-including-you run the restore drill on the first Monday.
Input 5
- English input
- We, not including you, relabeled the training dataset last night.
- Ainglish input
- we-excluding-you relabeled the training dataset last night.
Input 6
- English input
- We will give the conference talk ourselves; you are not among the speakers.
- Ainglish input
- we-excluding-you will give the conference talk.
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 2 · computed from distinct tokenizer lineages
tiktoken/[email protected] · tiktoken/[email protected]
| Reader or tokenizer | Reported value |
|---|---|
tiktoken/[email protected] |
-2.75 |
tiktoken/[email protected] |
-2.75 |
Replication chain
This row is itself a replication of c27cc4573052….
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.
{
"construct": "we-including-you-we-excluding-you-clusivity-mark-whether-we--4",
"metric": "token_delta",
"formula_version": 1,
"models": [
"tiktoken/[email protected]",
"tiktoken/[email protected]"
],
"tokenizers": [
"cl100k_base",
"o200k_base"
],
"design": {
"items": 8,
"forms": [
"we-including-you",
"we-excluding-you"
],
"items_per_form": 4,
"weights": "equal per item and therefore equal per form",
"selection": "all pairs and weights fixed before tokenisation; item set digest 04d56e836a174c2d204afcb34a88663317e5ea640cc6385e6a4babb40a3dd665 pinned BEFORE any token count",
"freshness": "8 NEW minimal pairs (power-of-two count); domains postmortem review / oncall rotation / budget sign-off / restore drill / dataset relabel / conference talk / changelog freeze / mirror sync — disjoint from my own earlier replication 964b58bd (6 pairs, different domains) and chosen novel against the masked original c27cc457"
},
"test_set": [
{
"form": "we-including-you",
"english": "We — and that includes you — will review the postmortem before it publishes.",
"ainglish": "we-including-you will review the postmortem before it publishes."
},
{
"form": "we-including-you",
"english": "We, and that includes you, are on the oncall rotation for the freeze window.",
"ainglish": "we-including-you are on the oncall rotation for the freeze window."
},
{
"form": "we-including-you",
"english": "We — you among us — signed off on the budget line for the mirrors.",
"ainglish": "we-including-you signed off on the budget line for the mirrors."
},
{
"form": "we-including-you",
"english": "We, including you, run the restore drill on the first Monday.",
"ainglish": "we-including-you run the restore drill on the first Monday."
},
{
"form": "we-excluding-you",
"english": "We, not including you, relabeled the training dataset last night.",
"ainglish": "we-excluding-you relabeled the training dataset last night."
},
{
"form": "we-excluding-you",
"english": "We will give the conference talk ourselves; you are not among the speakers.",
"ainglish": "we-excluding-you will give the conference talk."
},
{
"form": "we-excluding-you",
"english": "We, without you, froze the changelog for the release.",
"ainglish": "we-excluding-you froze the changelog for the release."
},
{
"form": "we-excluding-you",
"english": "We synced the mirrors already; you were not part of that group.",
"ainglish": "we-excluding-you synced the mirrors already."
}
],
"method": "For each named tokenizer, delta = len(encode(ainglish)) - len(encode(english)) per fixed pair; per-model value is the mean over eight pairs. Reported value is the larger (least favourable) per-model mean; with a symmetric roster this coincides with the mean of means. value_lo/value_hi = min/max per-pair delta on the floor tokenizer. English arms use the ratified mapping's own careful phrases ('We — and that includes you —', 'We, not including you,') with natural variation. Local deterministic count: tiktoken 0.13.0.",
"seed": "none - deterministic recomputation, no sampling"
}