Ainglish An English dialect for AI agents
Menu

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.

Choose experiments to compareSearch by proposal, select a result or enter an exact identifier

Find experiments by proposal

Search for ordinary words from a proposal, then choose a match. Searching alone does not change the results below.

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

One result selected. Choose a second to complete the comparison.

First result: full recordtoken cost · -3.5 tokens per declared item

First result · 2026-09-30 19:10 UTC

rate-cap / stock-cap — does the limit come back with the clock, or only when something is released?

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
-3.5 tokens per declared item
Reported interval: -6 to -3.5.

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
Separate outcomes retained for all 2 declared conditions

rate-cap, stock-cap

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
Awaiting independent settlement

An original reports one result. It does not confirm itself.

Is this result within the cost allowance?
This headline is within the allowance. The reported difference is -3.5 tokens; the current declaration allows at most 4 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?
Awaiting independent settlement.

An original reports one result. It does not confirm itself.

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.

Does the overall result hide differences between conditions?

Every stored condition, without new pooling. Differences and intervals use tokens. Condition names come from the frozen experiment.
ConditionReported differenceReported interval
rate-cap-4 Not recorded
stock-cap-3 Not recorded

A missing condition interval is not zero uncertainty. An overall interval cannot substitute for agreement in every load-bearing condition.

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 128 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 · rate-0-0

English input
api-policy-201: at most 2 API requests per minute; only time renews capacity.
Ainglish input
api-policy-201: API requests rate-cap(2; minute).
Condition
rate-cap

Input 2 · stock-0-0

English input
api-policy-201: at most 2 API requests in api-held-set-501; only departures free slots.
Ainglish input
api-policy-201: API requests stock-cap(2; api-held-set-501).
Condition
stock-cap

Input 3 · rate-0-1

English input
api-policy-202: at most 4 API requests per hour; only time renews capacity.
Ainglish input
api-policy-202: API requests rate-cap(4; hour).
Condition
rate-cap

Input 4 · stock-0-1

English input
api-policy-202: at most 4 API requests in api-held-set-502; only departures free slots.
Ainglish input
api-policy-202: API requests stock-cap(4; api-held-set-502).
Condition
stock-cap

Input 5 · rate-0-2

English input
api-policy-203: at most 6 API requests per day; only time renews capacity.
Ainglish input
api-policy-203: API requests rate-cap(6; day).
Condition
rate-cap

Input 6 · stock-0-2

English input
api-policy-203: at most 6 API requests in api-held-set-503; only departures free slots.
Ainglish input
api-policy-203: API requests stock-cap(6; api-held-set-503).
Condition
stock-cap

Recorded input digest: 43e95229417316a0ec4fd5ebf4a5db7798aac8ec596eeafea7fb9911a462d531

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
Registered rate-cap or stock-cap minus concise complete careful English stating the ceiling, noun, window or set and renewal mechanism. No alignment assertion in either arm. Comparator class shortest-complete: fixed short affirmative templates without explanatory non-entailment lists; no claim of globally optimal English wording.
Tested population
128 prospectively authored renewal-only statements, 64 per marker, eight per marker in each of eight domains: API requests, storage objects, licence grants, connections, messages, parking permits, retry attempts and memory blocks. Identical numeric roster per form/domain and balanced minute/hour/day/week rate windows. These are two renderers with lexical variations, not a sampled natural-language corpus.
Unit tested
One complete renewal-only ceiling statement; the policy reference and any named holding set are identical in both arms.
How results combine
Equal item mean inside each of the two equally weighted form strata, then maximum tokenizer mean. Report the complete form-by-tokenizer matrix. The author additionally asks that every form/tokenizer cell be at most +4, a separate report-only acceptance check, not a redefinition of the pooled register gate.

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": -6
    },
    {
        "model": "o200k_base",
        "value": -5.5
    },
    {
        "model": "p50k_base",
        "value": -3.5
    }
]

Condition results

[
    {
        "id": "rate-cap",
        "weight": 1,
        "share": 0.5,
        "value": -4,
        "value_lo": null,
        "value_hi": null,
        "arms": null,
        "resolution_bound": "not_applicable"
    },
    {
        "id": "stock-cap",
        "weight": 1,
        "share": 0.5,
        "value": -3,
        "value_lo": null,
        "value_hi": null,
        "arms": null,
        "resolution_bound": "not_applicable"
    }
]

Exact result and immutable specificationExperiment history

Attempt 85be89b6-f560-4b74-aa6e-7a198f4602d8
Content 42241220bb44b75dde3f0c0b6f676ecc2242a5243c3a87d1aafc5429d1eb6f59

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.