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?	on-purpose / by-accident — say whether an action you report was chosen or a slip2026-09-08 11:39 UTC	on-purpose / by-accident — say whether an action you report was chosen or a slip2026-09-09 12:00 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?	1.125 tokens per declared itemReported interval: 0.5 to 1.125Not yet counting in evidence decisions	-0.875 tokens per declared itemReported interval: -1.5 to -0.875Counts in current 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?	DisputedEligible replications disagree and this original does not hold a settlement majority.	Disagrees with the named originalThis eligible row adds one disagreement. An adverse or null direction is a valid result and remains visible.

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 · Same recorded value

First result

cl100k_base/o200k_base/p50k_base

Second result

cl100k_base/o200k_base/p50k_base

Unit tested · Same recorded value

First result

pair

Second result

pair

Named instruments (including order) · Same recorded value

First result

[
    "cl100k_base",
    "o200k_base",
    "p50k_base"
]

Second result

[
    "cl100k_base",
    "o200k_base",
    "p50k_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 · Same recorded value

First result

{
    "reducer": "least_favourable",
    "rule": "maximum tokenizer mean"
}

Second result

{
    "reducer": "least_favourable",
    "rule": "maximum tokenizer mean"
}

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 · 1.125 tokens per declared item

First result · 2026-09-08 11:39 UTC

on-purpose / by-accident — say whether an action you report was chosen or a slip

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: 1.125 tokens per declared item
Reported interval: 0.5 to 1.125.
More 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, p50k_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: Disputed
Eligible replications disagree and this original does not hold a settlement majority.

Is this result within the cost allowance?

This headline is within the allowance. The reported difference is 1.125 tokens; the current declaration allows at most 3 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?

Disputed.

Eligible replications disagree and this original does not hold 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 8 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: I deleted the stale release branch deliberately; the tags survive.
Ainglish input: I deleted the stale release branch on-purpose; the tags survive.

Input 2

English input: The rebase deleted the release branch, which I had not foreseen — restoring it from the reflog.
Ainglish input: The rebase deleted the release branch by-accident — restoring it from the reflog.

Input 3

English input: The migration was skipped deliberately: it needs the maintenance window.
Ainglish input: The migration was skipped on-purpose: it needs the maintenance window.

Input 4

English input: Handover: the cache was cleared by mistake during the test run, unforeseen; expect a cold start.
Ainglish input: Handover: the cache was cleared by-accident during the test run; expect a cold start.

Input 5

English input: I deployed the hotfix deliberately; the rollback plan is documented.
Ainglish input: I deployed the hotfix on-purpose; the rollback plan is documented.

Input 6

English input: The script deleted the production database by mistake; recovering from snapshot.
Ainglish input: The script deleted the production database by-accident; recovering from snapshot.

Recorded input digest: 0b11ee2289d74c77826ee552c7fce3b4be1a47cd96a6c29c50da2d0456e29b59

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/p50k_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.

Absolute arm results

Not recorded

Reader or tokenizer results

[
    {
        "model": "cl100k_base",
        "value": 0.5
    },
    {
        "model": "o200k_base",
        "value": 0.625
    },
    {
        "model": "p50k_base",
        "value": 1.125
    }
]

Condition results

Not recorded

Exact result and immutable specification Experiment history

Attempt 3e487cf0-3254-419d-b815-b459a761709f
Content 256a92882cc54e2347488c832bbcbd6171f8027482f4e1913217bebbe470ff24

Second result: full recordtoken cost · -0.875 tokens per declared item

Second result · 2026-09-09 12:00 UTC

on-purpose / by-accident — say whether an action you report was chosen or a slip

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: -0.875 tokens per declared item
Reported interval: -1.5 to -0.875.
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, p50k_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: Disagrees with the named original
This eligible row adds one disagreement. An adverse or null direction is a valid result and remains visible.

Is this result within the cost allowance?

This headline is within the allowance. The reported difference is -0.875 tokens; the current declaration allows at most 3 tokens.

Has the original estimate been independently reproduced?

Disagrees with the named original. This replication reports -0.875 tokens; the named original reported 1.125.

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.

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

256a92882cc54e2347488c832bbcbd6171f8027482f4e1913217bebbe470ff24

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.

Each result has its own input pages. Positions across the two studies do not imply matched cases.

Input 1

English input: I archived the obsolete build script deliberately; the release notes remain.
Ainglish input: I archived the obsolete build script on-purpose; the release notes remain.

Input 2

English input: The formatter overwrote the config comment, which I had not foreseen — retrieving it from version control.
Ainglish input: The formatter overwrote the config comment by-accident — retrieving it from version control.

Input 3

English input: The nightly export was paused deliberately: the storage node is under repair.
Ainglish input: The nightly export was paused on-purpose: the storage node is under repair.

Input 4

English input: Handover: the search index was reset by mistake during the smoke test, unforeseen; expect a rebuild.
Ainglish input: Handover: the search index was reset by-accident during the smoke test; expect a rebuild.

Input 5

English input: I rotated the staging certificate deliberately; the old certificate is retained.
Ainglish input: I rotated the staging certificate on-purpose; the old certificate is retained.

Input 6

English input: The cleanup task removed the fixture directory by mistake, an outcome its operator did not foresee; restoring the archive.
Ainglish input: The cleanup task removed the fixture directory by-accident; restoring the archive.

Recorded input digest: 5d38e82876c9e84ce9a85d6ccee251dc9209a077d4e29c750694090aea06851b

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/p50k_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

Not recorded

Reader or tokenizer results

[
    {
        "model": "cl100k_base",
        "value": -1.5
    },
    {
        "model": "o200k_base",
        "value": -1.375
    },
    {
        "model": "p50k_base",
        "value": -0.875
    }
]

Condition results

Not recorded

Exact result and immutable specification Experiment history

Attempt 9cb74303-51d1-48df-a515-e94eb64b6526
Content 4d82f2fda105bde0ecdb1bb985baabb729b47b403bf016ced3e91e97cac23e3c

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.