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
-19.25 tokens per declared item Reported interval: -19.3125 to -19.25.
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
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 16 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 accepted 1993-character, 2019-byte post: under the rival reading, a byte cap, it is rejected; under the reading I hold, a character cap, it is accepted; it was accepted, so it separates them.
Ainglish input
The accepted 1993-character, 2019-byte post tells-apart(a byte cap: reject | a character cap: accept).
Input 2
English input
The second read at 09:00Z returned the row after the edge cache expired: under the rival reading, an origin 404, it returns 404; under the reading I hold, a stale edge cache, it returns 200; it returned 200, so it separates them.
Ainglish input
The second read at 09:00Z returned the row after the edge cache expired tells-apart(an origin 404: 404 | a stale edge cache: 200).
Input 3
English input
The comment count caught up 70 seconds after the post count: under the rival reading, a single counter, there is no lag; under the reading I hold, a lagging replica, there is a lag; there was a lag, so it separates them.
Ainglish input
The comment count caught up 70 seconds after the post count tells-apart(a single counter: no lag | a lagging replica: a lag).
Input 4
English input
The 1983-character, 1985-byte body was accepted: under the rival reading, a byte cap, it is rejected; under the reading I hold, a character cap, it is accepted; it was accepted, so it separates them, and the test can only run on bodies whose character count and byte count fall on opposite sides of 2000.
Ainglish input
The 1983-character, 1985-byte body was accepted tells-apart(a byte cap: reject | a character cap: accept) over(bodies whose character count and byte count fall on opposite sides of 2000).
Input 5
English input
The 429 arrived on the sixth request within one second: under the rival reading, a per-key limit, that request returns 200; under the reading I hold, a per-address limit, it returns 429; it returned 429, so it separates them, and the test can only run where two keys share one address.
Ainglish input
The 429 arrived on the sixth request within one second tells-apart(a per-key limit: 200 | a per-address limit: 429) over(two keys sharing one address).
Input 6
English input
The 03:00Z retry succeeded without a new token: under the rival reading, token expiry, it succeeds; under the reading I hold, clock skew, it fails; it succeeded, so it separates them.
Ainglish input
The 03:00Z retry succeeded without a new token tells-apart(token expiry: succeed | clock skew: fail).
Recorded input digest: ecf759f2f7118f1bcb1b8e509794012ac64ead5b9a3ba6a3bc7d2c40e899acc4
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
Compared with
token_delta
Tested population
cl100k_base/o200k_base
Unit tested
pair
How results combine
maximum tokenizer mean
These are the study author’s declarations. A finding applies to this tested scope; this summary does not establish that another study is comparable.
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.
