Evidence

Compare two experiments

Check what changed in the question, wording or readers before comparing the numbers.

This view keeps both results separate. It does not calculate a combined score or decide whether they reproduce each other.

Change the selected experimentsSearch by proposal, select a result or enter an exact identifier

Find experiments by proposal

Choose two experiments by title, measurement and date. The choices include up to 50 newest public completed results, plus your current selections. Historical results stay labelled. For older records, use the evidence explorer or exact entry below.

Use an exact experiment identifier instead

A full attempt UUID entered here replaces the corresponding choice above. Content hashes are not result identifiers.

Clear comparison

Compare the essentials

Read each question across both results. A matching description is not proof that the studies are scientifically comparable.

Question	First result	Second result
Which proposal?	grader=graded2026-09-06 08:01 UTC	grader=graded2026-09-10 13:16 UTC
What was measured?	token costHow does the wording change tokenizer units for the declared tokenizer population?	token costHow does the wording change tokenizer units for the declared tokenizer population?
What was the result?	-16 tokens per declared itemReported interval: -17 to -16Counts in current evidence decisions	-16 tokens per declared itemReported interval: -17 to -16Not yet counting in evidence decisions
Compared with what?	English comparison not recorded as a structured labelCurrent tokenizer cost, not comprehension	English comparison not recorded as a structured labelCurrent tokenizer cost, not comprehension
Independently settled?	Confirmed by eligible settlementEligible fresh-input replications currently give this original a settlement majority.	Incommensurable pending repairA comparison key does not match, so the row cannot presently vote on settlement.

Inspect first result Inspect second result

What differs in the declared methods?

These are literal recorded values, not a compatibility test. Matching declarations do not prove equivalent inputs or fair scoring. A missing structured field may be described elsewhere in the immutable specification.

What was measured · Same recorded value

First result

token_delta

Second result

token_delta

English comparison declarations · Not recorded on either side

First result

Not recorded in this structured field

Second result

Not recorded in this structured field

Tested population · Only recorded on one side

First result

16 prospectively authored complete pairs, one per named subject; not random natural prose

Second result

Not recorded in this structured field

Unit tested · Only recorded on one side

First result

one complete scoped statement

Second result

Not recorded in this structured field

Named instruments (including order) · Same recorded value

First result

[
    "cl100k_base",
    "o200k_base"
]

Second result

[
    "cl100k_base",
    "o200k_base"
]

Reader exposure · Not recorded on either side

First result

Not recorded in this structured field

Second result

Not recorded in this structured field

Exposure window · Not recorded on either side

First result

Not recorded in this structured field

Second result

Not recorded in this structured field

How results combine · Only recorded on one side

First result

{
    "reducer": "least_favourable",
    "rule": "Equal pair mean in each cached encoding; maximum tokenizer mean across cl100k_base and o200k_base."
}

Second result

Not recorded in this structured field

Scoring declaration · Not recorded on either side

First result

Not recorded in this structured field

Second result

Not recorded in this structured field

Conditions (including order) · Not recorded on either side

First result

Not recorded in this structured field

Second result

Not recorded in this structured field

First result: full recordtoken cost · -16 tokens per declared item

First result · 2026-09-06 08:01 UTC

grader=graded

Counts in current evidence decisions. This row currently contributes to evidence decisions. Its direction is separate from whether the proposal is ready for adoption.

What was measured: token cost · token_delta
How does the wording change tokenizer units for the declared tokenizer population?
Reported result: -16 tokens per declared item
Reported interval: -17 to -16.
Fewer tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.
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: Confirmed by eligible settlement
Eligible fresh-input replications currently give this original a settlement majority.

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?

Confirmed by eligible settlement.

Eligible fresh-input replications currently give this original a settlement majority.

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.

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: Case G96-00: The balance checker has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G96-00: The balance checker is grader=graded.

Input 2

English input: Case G96-01: The library licence review has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G96-01: The library licence review is grader=graded.

Input 3

English input: Case G96-02: The attendance audit has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G96-02: The attendance audit is grader=graded.

Input 4

English input: Case G96-03: The voltage verifier has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G96-03: The voltage verifier is grader=graded.

Input 5

English input: Case G96-04: The translation evaluator has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G96-04: The translation evaluator is grader=graded.

Input 6

English input: Case G96-05: The risk rating review has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G96-05: The risk rating review is grader=graded.

Recorded input digest: 9928ed456920e11a60bd5fbfd203dcb0dc66a47fcd8248402e3db1a6f77f2eda

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: The current registered grader=graded definition applied to the same subject; no English-only mechanism, provenance, numbers or facts
Tested population: 16 prospectively authored complete pairs, one per named subject; not random natural prose
Unit tested: one complete scoped statement
How results combine: Equal pair mean in each cached encoding; maximum tokenizer mean across cl100k_base and o200k_base.

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.

Absolute arm results

Not recorded

Reader or tokenizer results

[
    {
        "model": "cl100k_base",
        "value": -16
    },
    {
        "model": "o200k_base",
        "value": -17
    }
]

Condition results

Not recorded

Exact result and immutable specification Experiment history

Attempt ed08aef0-9154-4efa-81c8-e9f2c4908aa3
Content d46360bbeb37d018476462d76ffc0f8faac580cebeea67867a74f424673fe7f2

Second result: full recordtoken cost · -16 tokens per declared item

Second result · 2026-09-10 13:16 UTC

grader=graded

Not yet counting in evidence decisions. This row remains available for assessment, but does not currently carry a counting evidence result.

What was measured: token cost · token_delta
How does the wording change tokenizer units for the declared tokenizer population?
Reported result: -16 tokens per declared item
Reported interval: -17 to -16.
Fewer tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.
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: Incommensurable pending repair
A comparison key does not match, so the row cannot presently vote on settlement.

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.

Has the original estimate been independently reproduced?

Incommensurable pending repair. This replication reports -16 tokens; the named original reported -16.

A comparison key does not match, so the row cannot presently vote on settlement.

Reproduction asks whether fresh-input findings agree under the settlement rule. It does not ask whether either value satisfies the cost allowance.

How can one check pass while the other does not?

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

Declared target content identity

d46360bbeb37d018476462d76ffc0f8faac580cebeea67867a74f424673fe7f2

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: Case G97-00: The latency benchmark review has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G97-00: The latency benchmark review is grader=graded.

Input 2

English input: Case G97-01: The invoice matcher audit has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G97-01: The invoice matcher audit is grader=graded.

Input 3

English input: Case G97-02: The access log inspector has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G97-02: The access log inspector is grader=graded.

Input 4

English input: Case G97-03: The payload checksum test has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G97-03: The payload checksum test is grader=graded.

Input 5

English input: Case G97-04: The roster eligibility check has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G97-04: The roster eligibility check is grader=graded.

Input 6

English input: Case G97-05: The signal calibration review has an evaluator that shares state with the evaluated party; a pass certifies agreement with itself, not correctness.
Ainglish input: Case G97-05: The signal calibration review is grader=graded.

Recorded input digest: fe9afcb71c482f951a12543edda92fcc533f73f039be4034cbabbcbfb0cd3dad

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

No structured study scope is declared here. Inspect the immutable manifest; do not infer a comparator or population from the headline.

Absolute arm results

Not recorded

Reader or tokenizer results

[
    {
        "model": "cl100k_base",
        "value": -16
    },
    {
        "model": "o200k_base",
        "value": -17
    }
]

Condition results

Not recorded

Exact result and immutable specification Experiment history

Attempt 5e58c0df-047d-45af-ae34-c328674c9dbf
Content 654551c9606b65e92215314875fc7b4cb621fa8913a251ddffd5d8c6ccfd3985

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.