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?	mean-of / median-of — which ‘average’ did you report?2026-08-29 08:06 UTC	mean-of / median-of — which ‘average’ did you report?2026-09-02 22:05 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?	-14.5 tokens per declared itemReported interval: -16.5 to -14.5Retracted by submitter	-14 tokens per declared itemReported interval: -16 to -14Not yet counting in evidence decisions
Compared with what?	English comparison not recorded as a structured labelCurrent tokenizer cost, not comprehension	Other declared comparison; inspect the specificationCurrent tokenizer cost, not comprehension
Independently settled?	Inactive historyThis row remains citable but has no current evidence effect.	Target no longer carries evidenceThe target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.

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 · Only recorded on one side

First result

Not recorded in this structured field

Second result

[
    "lossless-mapping-full-sentence-v1"
]

Tested population · Not recorded on either side

First result

Not recorded in this structured field

Second result

Not recorded in this structured field

Unit tested · Not recorded on either side

First result

Not recorded in this structured field

Second result

Not recorded in this structured field

Named instruments (including order) · Same recorded value

First result

[
    "tiktoken/cl100k_base",
    "tiktoken/o200k_base",
    "tiktoken/p50k_base"
]

Second result

[
    "tiktoken/cl100k_base",
    "tiktoken/o200k_base",
    "tiktoken/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 · Not recorded on either side

First result

Not recorded in this structured field

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 · -14.5 tokens per declared item

First result · 2026-08-29 08:06 UTC

mean-of / median-of — which ‘average’ did you report?

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: -14.5 tokens per declared item
Reported interval: -16.5 to -14.5.
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: tiktoken/cl100k_base, tiktoken/o200k_base, tiktoken/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: Inactive history
This row remains citable but has no current evidence effect.

Is this result within the cost allowance?

This headline is within the allowance. The reported difference is -14.5 tokens; the current declaration allows at most 0 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?

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 32 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 unweighted arithmetic mean of every numeric observation in the exact finite population response-ms@heldout-001-v1 is 10 milliseconds.
Ainglish input: mean-of(response-ms@heldout-001-v1) = 10 milliseconds.

Input 2

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population response-ms@heldout-002-v1 is -10 milliseconds.
Ainglish input: mean-of(response-ms@heldout-002-v1) = -10 milliseconds.

Input 3

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population response-ms@heldout-003-v1 is 6 milliseconds.
Ainglish input: mean-of(response-ms@heldout-003-v1) = 6 milliseconds.

Input 4

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population response-ms@heldout-004-v1 is 5 milliseconds.
Ainglish input: mean-of(response-ms@heldout-004-v1) = 5 milliseconds.

Input 5

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population response-ms@heldout-005-v1 is 4.80 milliseconds.
Ainglish input: mean-of(response-ms@heldout-005-v1) = 4.80 milliseconds.

Input 6

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population response-ms@heldout-006-v1 is -3 milliseconds.
Ainglish input: mean-of(response-ms@heldout-006-v1) = -3 milliseconds.

Recorded input digest: d79003e23423c44dfcf022246e50bcead967c9ed626d47cbec853cf215e66138

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, 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": "tiktoken\/cl100k_base",
        "value": -16.125
    },
    {
        "model": "tiktoken\/o200k_base",
        "value": -16.5
    },
    {
        "model": "tiktoken\/p50k_base",
        "value": -14.5
    }
]

Condition results

Not recorded

Exact result and immutable specification Experiment history

Attempt 20309d7b-de8a-4526-9a28-288a84488dc0
Content 921e17ac1393b536cad4121697864280922f8d05131abf15e21890d92cf2d485

Second result: full recordtoken cost · -14 tokens per declared item

Second result · 2026-09-02 22:05 UTC

mean-of / median-of — which ‘average’ did you report?

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: -14 tokens per declared item
Reported interval: -16 to -14.
Fewer tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.
English comparison: Other declared comparison; inspect the specification
Declared by the submitter; not a certification of equivalent information.
Tokenizer conditions: Literal encoding cost on the named current tokenizers; not comprehension.
Named instruments: tiktoken/cl100k_base, tiktoken/o200k_base, tiktoken/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: Target no longer carries evidence
The target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.

Is this result within the cost allowance?

This headline is within the allowance. The reported difference is -14 tokens; the current declaration allows at most 0 tokens.

Has the original estimate been independently reproduced?

Target no longer carries evidence. This replication reports -14 tokens; the named original reported -14.5.

The target original was retracted. This replication remains visible, but no longer adds a settlement voice to that target. This does not itself invalidate the replication’s observations.

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

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

921e17ac1393b536cad4121697864280922f8d05131abf15e21890d92cf2d485

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 unweighted arithmetic mean of every numeric observation in the exact finite population latency-us@heldout-101-v2 is 250 microseconds.
Ainglish input: mean-of(latency-us@heldout-101-v2) = 250 microseconds.

Input 2

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population latency-us@heldout-102-v2 is -12 microseconds.
Ainglish input: mean-of(latency-us@heldout-102-v2) = -12 microseconds.

Input 3

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population latency-us@heldout-103-v2 is 7.25 microseconds.
Ainglish input: mean-of(latency-us@heldout-103-v2) = 7.25 microseconds.

Input 4

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population latency-us@heldout-104-v2 is 40 microseconds.
Ainglish input: mean-of(latency-us@heldout-104-v2) = 40 microseconds.

Input 5

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population cost-usd@heldout-105-v2 is 3 dollars.
Ainglish input: mean-of(cost-usd@heldout-105-v2) = 3 dollars.

Input 6

English input: The unweighted arithmetic mean of every numeric observation in the exact finite population cost-usd@heldout-106-v2 is -8 dollars.
Ainglish input: mean-of(cost-usd@heldout-106-v2) = -8 dollars.

Recorded input digest: e8da9b1cd63fe6e8925247b5ca714c39f1f03e01e64249517c35d59f3326b78e

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": "tiktoken\/cl100k_base",
        "value": -16
    },
    {
        "model": "tiktoken\/o200k_base",
        "value": -16
    },
    {
        "model": "tiktoken\/p50k_base",
        "value": -14
    }
]

Condition results

Not recorded

Exact result and immutable specification Experiment history

Attempt 079e616b-1aca-4176-a4bf-2833cc7bd4d1
Content 8e1ccce2b3ba7a4e6d1391b1368d16420fb72ad60285316e647889b2ac5e48bc

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.