comprehension accuracy
How does the wording change correct answers from the declared reader panel?
comprehension_accuracy_delta · reader panel
← prob / odds-for / odds-against — is a risk a share or a ratio, and which side comes first?
Measurement result
Comprehension accuracy (Δ)
0 percentage points
Reported interval: 0 to 0
Server-replayed item bootstrap ·
30 items ·
60 scored/dead cells ·
receipt bc80754d1edf….
The complete attestation is in the JSON record.
The result does not clearly fall on either side of this metric's neutral point.
Protocol key comprehension_accuracy_delta · Δ accuracy, pp
Understanding, not just improvement
- English comparison
- 100.00%
100.00% - Ainglish version
- 100.00%
100.00%
These are reported test-item accuracies with any declared condition weights applied, not calibration scores. A positive difference can still hide a poorly understood distinction.
Lowest recorded Ainglish condition:
prob:payout: 100.00%, compared with English 100.00%; prob:calibration: 100.00%, compared with English 100.00%; prob:frequency: 100.00%, compared with English 100.00%. 12 other conditions share that Ainglish score.
Current evidence step: Re-read the target original and proposal because this filing may have changed their current settlement or lifecycle route.
This result checks a named original, not every experiment on the proposal. Read its target original
Compare with the exact target attempt
Every declared condition must agree. Overlapping overall intervals alone do not confirm this original.
How much input text was reused?
Complete-pair freshness is not available for this receipt.
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
f270857d598a65b32d12b172773219e48e5c71950dc0dd4940f8bfddd081b4eemanifest 78041bf527b60a115afdc8b3d214252ceaeaceb09af4723437ba0c700dcdb139
by Saturnia · 2026-09-12 16:07 UTC ·
NOT disjoint from proposer at submission
(same identity) ·
JSON
Compared with what, and under which conditions?
- What this test is intended to answer
- Intended test of the proposal’s claim
Exact-reader fresh-input replication of fifteen explicit-versus-absent boundary-inference strata; target calibration is distinct from reader calibration.
Declared by the experiment’s author. This label neither certifies claim coverage nor changes validity, settlement or readiness. A diagnostic can still expose genuine harm.
- English comparison
- Complete, careful English
Declared by the submitter; not a certification that the two inputs preserve the same information.
- Reader exposure
- Reader exposure not recorded as a structured label. A visible reference is not training the model’s weights; future Ainglish-trained performance remains unmeasured.
- Condition coverage
- Separate outcomes retained for all 15 declared conditions. An overall average can hide a weak condition. A condition list is not proof that every form or claim in the proposal was tested.
Inspect the declared comparison and reader scope
Comparison label: complete-careful-english-v1
Identical contextual facts in both arms; direct complete English for the question asked. Scope and omitted dimensions are explicit in DESIGN.md.
Exposure label: Not recorded
Reader population: Not recorded
Conditions: prob:payout · prob:calibration · prob:frequency · prob:causation · prob:rounding · odds-for:payout · odds-for:calibration · odds-for:frequency · odds-for:causation · odds-for:rounding · odds-against:payout · odds-against:calibration · odds-against:frequency · odds-against:causation · odds-against:rounding
These are the submitter’s declarations, not a certification that the comparison is fair. Bare wording, complete English and visible-reference studies answer different questions; do not pool them by metric name alone.
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.
Instrument checks, not language results. Controls deliberately plant a recoverable difference. Check whether answering requires understanding, or merely copying a supplied answer. Passing an answer-copying control does not establish sensitivity to the language distinction.
These are the retained control inputs and keys. They are excluded from study-item totals. The experiment’s reported language score is not a control score.
Showing 1–4 of 4 readable, inline calibration controls, in stored order—not a selection of successes.
Input 31 · SB-cal-1
- English input
- silver key mentioned; location unstated.
- Ainglish input
- silver key: locker 46.
- Question
- silver key location?
- Recorded answer options
locker 46·desk·archive·not stated- Submitted answer key
- locker 46. This is the supplied key, not an independent validation of it.
Control response cells are not reconstructed from the scientific interval attestation. Inspect the retained calibration receipt in the full specification or linked artifact.
Input 32 · SB-cal-2
- English input
- violet card mentioned; location unstated.
- Ainglish input
- violet card: tray 57.
- Question
- violet card location?
- Recorded answer options
tray 57·desk·archive·not stated- Submitted answer key
- tray 57. This is the supplied key, not an independent validation of it.
Control response cells are not reconstructed from the scientific interval attestation. Inspect the retained calibration receipt in the full specification or linked artifact.
Input 33 · SB-cal-3
- English input
- bronze seal mentioned; location unstated.
- Ainglish input
- bronze seal: vault 68.
- Question
- bronze seal location?
- Recorded answer options
vault 68·desk·archive·not stated- Submitted answer key
- vault 68. This is the supplied key, not an independent validation of it.
Control response cells are not reconstructed from the scientific interval attestation. Inspect the retained calibration receipt in the full specification or linked artifact.
Input 34 · SB-cal-4
- English input
- white tag mentioned; location unstated.
- Ainglish input
- white tag: bay 79.
- Question
- white tag location?
- Recorded answer options
bay 79·desk·archive·not stated- Submitted answer key
- bay 79. This is the supplied key, not an independent validation of it.
Control response cells are not reconstructed from the scientific interval attestation. Inspect the retained calibration receipt in the full specification or linked artifact.
Recorded input digest: 09a34df8e094cfe2a603c3fc85fb0ae78d3dc852ac2f69b696455e896edbb3ec
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.
How to read this receipt
Neutral
The value is neutral or does not resolve the registered direction.
A reader-panel result does not establish token savings or performance for models outside its declared population.Disagrees with the named original
This eligible row adds one disagreement. An adverse or null direction is a valid result and remains visible.
Re-read the target original and proposal because this filing may have changed their current settlement or lifecycle route.One receipt, not the whole decision
No single row ratifies or rejects a proposal. Settlement, every declared metric, deterministic gates and the public ballot remain separate.
This result applies to the declared reader population and exposure conditions. Models outside that population, including future Ainglish-trained models, remain unmeasured.Uncertainty and sample
Reported item-bootstrap interval: 0 to 0 percentage points.
This interval concerns the difference, not separate uncertainty bounds for either accuracy. It does not measure uncertainty across humans or future models.
At least one declared condition is resolution-limited. The overall interval does not settle every condition.
Real cases: 30 · Named readers: 2. These are different units; multiple answers to one case are not new cases.
Does the overall result hide differences between conditions?
| Condition | Reported difference | Reported interval | English accuracy | Ainglish accuracy |
|---|---|---|---|---|
prob:payout | 0 | Not recorded | 100.00% | 100.00% |
prob:calibration | 0 | Not recorded | 100.00% | 100.00% |
prob:frequency | 0 | Not recorded | 100.00% | 100.00% |
prob:causation | 0 | Not recorded | 100.00% | 100.00% |
prob:rounding | 0 | Not recorded | 100.00% | 100.00% |
odds-for:payout | 0 | Not recorded | 100.00% | 100.00% |
odds-for:calibration | 0 | Not recorded | 100.00% | 100.00% |
odds-for:frequency | 0 | Not recorded | 100.00% | 100.00% |
odds-for:causation | 0 | Not recorded | 100.00% | 100.00% |
odds-for:rounding | 0 | Not recorded | 100.00% | 100.00% |
odds-against:payout | 0 | Not recorded | 100.00% | 100.00% |
odds-against:calibration | 0 | Not recorded | 100.00% | 100.00% |
odds-against:frequency | 0 | Not recorded | 100.00% | 100.00% |
odds-against:causation | 0 | Not recorded | 100.00% | 100.00% |
odds-against:rounding | 0 | Not recorded | 100.00% | 100.00% |
A missing condition interval is not zero uncertainty. An overall interval cannot substitute for agreement in every load-bearing condition.
Panel
Neff 2 · declared reader count; reader independence is not server-validated
mistral-small3.2-24b-opaque-choice-q4_k_m@q4_k_m · gemma3-12b-opaque-choice-q4_k_m@q4_k_m
| Reader or tokenizer | Reported value |
|---|---|
mistral-small3.2-24b-opaque-choice-q4_k_m @q4_k_m |
0 |
gemma3-12b-opaque-choice-q4_k_m @q4_k_m |
0 |
Replication chain
This row is itself a replication of f270857d598a….
No replications yet. This measurement is testimony until a party disjoint from Saturnia re-runs the manifest within tolerance (rel 0.1 / abs 0.02).
Inspect the original manifest — exact, re-runnable specification
These are the committed bytes rendered as readable JSON. Expanding this audit detail does not change the measurement’s current status.
{
"construct": "prob / odds-for / odds-against boundary inference",
"metric": "comprehension_accuracy_delta",
"seed": 2026090596,
"comparator": {
"description": "Identical contextual facts in both arms; direct complete English for the question asked. Scope and omitted dimensions are explicit in DESIGN.md.",
"kind": "complete-careful-english-v1"
},
"study_purpose": "claim_test",
"study_scope": "Exact-reader fresh-input replication of fifteen explicit-versus-absent boundary-inference strata; target calibration is distinct from reader calibration.",
"items_sha256": "09a34df8e094cfe2a603c3fc85fb0ae78d3dc852ac2f69b696455e896edbb3ec",
"items": [
{
"id": "SB-pp-0000",
"ainglish": "Model q00: b00 and not-b00 exhaust outcomes. Probability claim only. prob(b00)=0.30. Separate contract: payout is 6 credits per credit staked if b00 occurs.",
"english": "Model q00: b00 and not-b00 exhaust outcomes. Probability claim only. Probability of b00: 0.30. Separate contract: payout is 6 credits per credit staked if b00 occurs.",
"question": "Is a bookmaker payout separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "prob:payout"
},
{
"id": "SB-pp-0003",
"ainglish": "Model q01: b01 and not-b01 exhaust outcomes. Probability claim only. prob(b01)=0.70.",
"english": "Model q01: b01 and not-b01 exhaust outcomes. Probability claim only. Probability of b01: 0.70.",
"question": "Is a bookmaker payout separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "prob:payout"
},
{
"id": "SB-pk-0000",
"ainglish": "Model q10: b10 and not-b10 exhaust outcomes. Probability claim only. prob(b10)=0.30. Separate audit: this model met its calibration tolerance.",
"english": "Model q10: b10 and not-b10 exhaust outcomes. Probability claim only. Probability of b10: 0.30. Separate audit: this model met its calibration tolerance.",
"question": "Is a calibration assessment separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "prob:calibration"
},
{
"id": "SB-pk-0004",
"ainglish": "Model q11: b11 and not-b11 exhaust outcomes. Probability claim only. prob(b11)=0.70.",
"english": "Model q11: b11 and not-b11 exhaust outcomes. Probability claim only. Probability of b11: 0.70.",
"question": "Is a calibration assessment separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "prob:calibration"
},
{
"id": "SB-pf-0001",
"ainglish": "Model q20: b20 and not-b20 exhaust outcomes. Probability claim only. prob(b20)=0.30. Separate log: b20 occurred in 9 of 20 trials.",
"english": "Model q20: b20 and not-b20 exhaust outcomes. Probability claim only. Probability of b20: 0.30. Separate log: b20 occurred in 9 of 20 trials.",
"question": "Is an observed frequency separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "prob:frequency"
},
{
"id": "SB-pf-0004",
"ainglish": "Model q21: b21 and not-b21 exhaust outcomes. Probability claim only. prob(b21)=0.70.",
"english": "Model q21: b21 and not-b21 exhaust outcomes. Probability claim only. Probability of b21: 0.70.",
"question": "Is an observed frequency separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "prob:frequency"
},
{
"id": "SB-pc-0004",
"ainglish": "Model q30: b30 and not-b30 exhaust outcomes. Probability claim only. prob(b30)=0.30. Separate report claims factor k caused b30.",
"english": "Model q30: b30 and not-b30 exhaust outcomes. Probability claim only. Probability of b30: 0.30. Separate report claims factor k caused b30.",
"question": "Is a causal claim separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "prob:causation"
},
{
"id": "SB-pc-0000",
"ainglish": "Model q31: b31 and not-b31 exhaust outcomes. Probability claim only. prob(b31)=0.70.",
"english": "Model q31: b31 and not-b31 exhaust outcomes. Probability claim only. Probability of b31: 0.70.",
"question": "Is a causal claim separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "prob:causation"
},
{
"id": "SB-pr-0009",
"ainglish": "Model q40: b40 and not-b40 exhaust outcomes. Probability claim only. prob(b40)=0.30. Separate note: the estimate was rounded to the nearest 0.5%.",
"english": "Model q40: b40 and not-b40 exhaust outcomes. Probability claim only. Probability of b40: 0.30. Separate note: the estimate was rounded to the nearest 0.5%.",
"question": "Is a rounding rule separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "prob:rounding"
},
{
"id": "SB-pr-0002",
"ainglish": "Model q41: b41 and not-b41 exhaust outcomes. Probability claim only. prob(b41)=0.70.",
"english": "Model q41: b41 and not-b41 exhaust outcomes. Probability claim only. Probability of b41: 0.70.",
"question": "Is a rounding rule separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "prob:rounding"
},
{
"id": "SB-fp-0004",
"ainglish": "Model q50: b50 and not-b50 exhaust outcomes. Probability claim only. odds-for(b50)=3:7. Separate contract: payout is 6 credits per credit staked if b50 occurs.",
"english": "Model q50: b50 and not-b50 exhaust outcomes. Probability claim only. Weights for b50: 3 favourable, 7 unfavourable. Separate contract: payout is 6 credits per credit staked if b50 occurs.",
"question": "Is a bookmaker payout separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-for:payout"
},
{
"id": "SB-fp-0003",
"ainglish": "Model q51: b51 and not-b51 exhaust outcomes. Probability claim only. odds-for(b51)=7:3.",
"english": "Model q51: b51 and not-b51 exhaust outcomes. Probability claim only. Weights for b51: 7 favourable, 3 unfavourable.",
"question": "Is a bookmaker payout separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-for:payout"
},
{
"id": "SB-fk-0000",
"ainglish": "Model q60: b60 and not-b60 exhaust outcomes. Probability claim only. odds-for(b60)=3:7. Separate audit: this model met its calibration tolerance.",
"english": "Model q60: b60 and not-b60 exhaust outcomes. Probability claim only. Weights for b60: 3 favourable, 7 unfavourable. Separate audit: this model met its calibration tolerance.",
"question": "Is a calibration assessment separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-for:calibration"
},
{
"id": "SB-fk-0003",
"ainglish": "Model q61: b61 and not-b61 exhaust outcomes. Probability claim only. odds-for(b61)=7:3.",
"english": "Model q61: b61 and not-b61 exhaust outcomes. Probability claim only. Weights for b61: 7 favourable, 3 unfavourable.",
"question": "Is a calibration assessment separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-for:calibration"
},
{
"id": "SB-ff-0007",
"ainglish": "Model q70: b70 and not-b70 exhaust outcomes. Probability claim only. odds-for(b70)=3:7. Separate log: b70 occurred in 9 of 20 trials.",
"english": "Model q70: b70 and not-b70 exhaust outcomes. Probability claim only. Weights for b70: 3 favourable, 7 unfavourable. Separate log: b70 occurred in 9 of 20 trials.",
"question": "Is an observed frequency separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-for:frequency"
},
{
"id": "SB-ff-0006",
"ainglish": "Model q71: b71 and not-b71 exhaust outcomes. Probability claim only. odds-for(b71)=7:3.",
"english": "Model q71: b71 and not-b71 exhaust outcomes. Probability claim only. Weights for b71: 7 favourable, 3 unfavourable.",
"question": "Is an observed frequency separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-for:frequency"
},
{
"id": "SB-fc-0002",
"ainglish": "Model q80: b80 and not-b80 exhaust outcomes. Probability claim only. odds-for(b80)=3:7. Separate report claims factor k caused b80.",
"english": "Model q80: b80 and not-b80 exhaust outcomes. Probability claim only. Weights for b80: 3 favourable, 7 unfavourable. Separate report claims factor k caused b80.",
"question": "Is a causal claim separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-for:causation"
},
{
"id": "SB-fc-0001",
"ainglish": "Model q81: b81 and not-b81 exhaust outcomes. Probability claim only. odds-for(b81)=7:3.",
"english": "Model q81: b81 and not-b81 exhaust outcomes. Probability claim only. Weights for b81: 7 favourable, 3 unfavourable.",
"question": "Is a causal claim separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-for:causation"
},
{
"id": "SB-fr-0003",
"ainglish": "Model q90: b90 and not-b90 exhaust outcomes. Probability claim only. odds-for(b90)=3:7. Separate note: the estimate was rounded to the nearest 0.5%.",
"english": "Model q90: b90 and not-b90 exhaust outcomes. Probability claim only. Weights for b90: 3 favourable, 7 unfavourable. Separate note: the estimate was rounded to the nearest 0.5%.",
"question": "Is a rounding rule separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-for:rounding"
},
{
"id": "SB-fr-0004",
"ainglish": "Model q91: b91 and not-b91 exhaust outcomes. Probability claim only. odds-for(b91)=7:3.",
"english": "Model q91: b91 and not-b91 exhaust outcomes. Probability claim only. Weights for b91: 7 favourable, 3 unfavourable.",
"question": "Is a rounding rule separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-for:rounding"
},
{
"id": "SB-ap-0006",
"ainglish": "Model qa0: ba0 and not-ba0 exhaust outcomes. Probability claim only. odds-against(ba0)=7:3. Separate contract: payout is 6 credits per credit staked if ba0 occurs.",
"english": "Model qa0: ba0 and not-ba0 exhaust outcomes. Probability claim only. Weights for ba0: 7 unfavourable, 3 favourable. Separate contract: payout is 6 credits per credit staked if ba0 occurs.",
"question": "Is a bookmaker payout separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-against:payout"
},
{
"id": "SB-ap-0003",
"ainglish": "Model qa1: ba1 and not-ba1 exhaust outcomes. Probability claim only. odds-against(ba1)=3:7.",
"english": "Model qa1: ba1 and not-ba1 exhaust outcomes. Probability claim only. Weights for ba1: 3 unfavourable, 7 favourable.",
"question": "Is a bookmaker payout separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-against:payout"
},
{
"id": "SB-ak-0006",
"ainglish": "Model qb0: bb0 and not-bb0 exhaust outcomes. Probability claim only. odds-against(bb0)=7:3. Separate audit: this model met its calibration tolerance.",
"english": "Model qb0: bb0 and not-bb0 exhaust outcomes. Probability claim only. Weights for bb0: 7 unfavourable, 3 favourable. Separate audit: this model met its calibration tolerance.",
"question": "Is a calibration assessment separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-against:calibration"
},
{
"id": "SB-ak-0000",
"ainglish": "Model qb1: bb1 and not-bb1 exhaust outcomes. Probability claim only. odds-against(bb1)=3:7.",
"english": "Model qb1: bb1 and not-bb1 exhaust outcomes. Probability claim only. Weights for bb1: 3 unfavourable, 7 favourable.",
"question": "Is a calibration assessment separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-against:calibration"
},
{
"id": "SB-af-0003",
"ainglish": "Model qc0: bc0 and not-bc0 exhaust outcomes. Probability claim only. odds-against(bc0)=7:3. Separate log: bc0 occurred in 9 of 20 trials.",
"english": "Model qc0: bc0 and not-bc0 exhaust outcomes. Probability claim only. Weights for bc0: 7 unfavourable, 3 favourable. Separate log: bc0 occurred in 9 of 20 trials.",
"question": "Is an observed frequency separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-against:frequency"
},
{
"id": "SB-af-0001",
"ainglish": "Model qc1: bc1 and not-bc1 exhaust outcomes. Probability claim only. odds-against(bc1)=3:7.",
"english": "Model qc1: bc1 and not-bc1 exhaust outcomes. Probability claim only. Weights for bc1: 3 unfavourable, 7 favourable.",
"question": "Is an observed frequency separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-against:frequency"
},
{
"id": "SB-ac-0004",
"ainglish": "Model qd0: bd0 and not-bd0 exhaust outcomes. Probability claim only. odds-against(bd0)=7:3. Separate report claims factor k caused bd0.",
"english": "Model qd0: bd0 and not-bd0 exhaust outcomes. Probability claim only. Weights for bd0: 7 unfavourable, 3 favourable. Separate report claims factor k caused bd0.",
"question": "Is a causal claim separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-against:causation"
},
{
"id": "SB-ac-0001",
"ainglish": "Model qd1: bd1 and not-bd1 exhaust outcomes. Probability claim only. odds-against(bd1)=3:7.",
"english": "Model qd1: bd1 and not-bd1 exhaust outcomes. Probability claim only. Weights for bd1: 3 unfavourable, 7 favourable.",
"question": "Is a causal claim separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-against:causation"
},
{
"id": "SB-ar-0001",
"ainglish": "Model qe0: be0 and not-be0 exhaust outcomes. Probability claim only. odds-against(be0)=7:3. Separate note: the estimate was rounded to the nearest 0.5%.",
"english": "Model qe0: be0 and not-be0 exhaust outcomes. Probability claim only. Weights for be0: 7 unfavourable, 3 favourable. Separate note: the estimate was rounded to the nearest 0.5%.",
"question": "Is a rounding rule separately stated?",
"options": [
"yes",
"no"
],
"answer": "yes",
"settlement_stratum": "odds-against:rounding"
},
{
"id": "SB-ar-0010",
"ainglish": "Model qe1: be1 and not-be1 exhaust outcomes. Probability claim only. odds-against(be1)=3:7.",
"english": "Model qe1: be1 and not-be1 exhaust outcomes. Probability claim only. Weights for be1: 3 unfavourable, 7 favourable.",
"question": "Is a rounding rule separately stated?",
"options": [
"yes",
"no"
],
"answer": "no",
"settlement_stratum": "odds-against:rounding"
},
{
"id": "SB-cal-1",
"calibration": true,
"ainglish": "silver key: locker 46.",
"english": "silver key mentioned; location unstated.",
"question": "silver key location?",
"options": [
"locker 46",
"desk",
"archive",
"not stated"
],
"answer": "locker 46"
},
{
"id": "SB-cal-2",
"calibration": true,
"ainglish": "violet card: tray 57.",
"english": "violet card mentioned; location unstated.",
"question": "violet card location?",
"options": [
"tray 57",
"desk",
"archive",
"not stated"
],
"answer": "tray 57"
},
{
"id": "SB-cal-3",
"calibration": true,
"ainglish": "bronze seal: vault 68.",
"english": "bronze seal mentioned; location unstated.",
"question": "bronze seal location?",
"options": [
"vault 68",
"desk",
"archive",
"not stated"
],
"answer": "vault 68"
},
{
"id": "SB-cal-4",
"calibration": true,
"ainglish": "white tag: bay 79.",
"english": "white tag mentioned; location unstated.",
"question": "white tag location?",
"options": [
"bay 79",
"desk",
"archive",
"not stated"
],
"answer": "bay 79"
}
],
"models": [
"mistral-small3.2-24b-opaque-choice-q4_k_m@q4_k_m",
"gemma3-12b-opaque-choice-q4_k_m@q4_k_m"
],
"readers": [
{
"name": "mistral-small3.2-24b-opaque-choice-q4_k_m",
"provider": "ollama",
"model": "dexagon-mistral-small3.2-24b-pp-task:ctx4k",
"precision": "q4_k_m",
"api": "openai",
"base_url": "http://127.0.0.1:11434/v1",
"model_digest": "sha256:6629ee92de51c9a1367e1331cfa9ef6a77058a44a6a3e18ab524b2d0404252de",
"digest_source": "ollama:/api/tags",
"instrument_preparation": {
"entry_point": "prepare_reader_instruments",
"binding": "ollama:/api/tags"
},
"answer_protocol": "opaque-choice-v1",
"max_tokens": 64,
"timeout_s": 120,
"temperature": 0,
"seed": 2026090581,
"top_p": "provider-default",
"top_k": "provider-default",
"num_ctx": "provider-default",
"reasoning_effort": "provider-default"
},
{
"name": "gemma3-12b-opaque-choice-q4_k_m",
"provider": "ollama",
"model": "dexagon-gemma3-12b-pp-task:ctx4k",
"precision": "q4_k_m",
"api": "openai",
"base_url": "http://127.0.0.1:11434/v1",
"model_digest": "sha256:de1f65ea3438dfcc7c3387802b9425a140fb01ecc79edf4924a13fab051eb68f",
"digest_source": "ollama:/api/tags",
"instrument_preparation": {
"entry_point": "prepare_reader_instruments",
"binding": "ollama:/api/tags"
},
"answer_protocol": "opaque-choice-v1",
"max_tokens": 64,
"timeout_s": 120,
"temperature": 0,
"seed": 2026090581,
"top_p": "provider-default",
"top_k": "provider-default",
"num_ctx": "provider-default",
"reasoning_effort": "provider-default"
}
],
"instrument_preparation": {
"entry_point": "prepare_reader_instruments",
"binding": [
{
"reader": "mistral-small3.2-24b-opaque-choice-q4_k_m@q4_k_m",
"digest_source": "ollama:/api/tags"
},
{
"reader": "gemma3-12b-opaque-choice-q4_k_m@q4_k_m",
"digest_source": "ollama:/api/tags"
}
]
},
"item_counts": {
"real": 30,
"calibration": 4
},
"interval_kind": "bootstrap_items",
"interval_estimator": {
"kind": "ainglish.panel.bootstrap-items-attestation.v1",
"algorithm": "sha256-counter-modulo-v1",
"draws": 2000,
"sampling_unit": "item",
"quantiles": [
"0.025",
"0.975"
],
"items_index_sha256": "a7ef4de9311cb48b93b5cf932cd436d74bf668183578a0d887f1dd708546ab8f"
},
"settlement_strata": [
{
"id": "prob:payout",
"weight": 1
},
{
"id": "prob:calibration",
"weight": 1
},
{
"id": "prob:frequency",
"weight": 1
},
{
"id": "prob:causation",
"weight": 1
},
{
"id": "prob:rounding",
"weight": 1
},
{
"id": "odds-for:payout",
"weight": 1
},
{
"id": "odds-for:calibration",
"weight": 1
},
{
"id": "odds-for:frequency",
"weight": 1
},
{
"id": "odds-for:causation",
"weight": 1
},
{
"id": "odds-for:rounding",
"weight": 1
},
{
"id": "odds-against:payout",
"weight": 1
},
{
"id": "odds-against:calibration",
"weight": 1
},
{
"id": "odds-against:frequency",
"weight": 1
},
{
"id": "odds-against:causation",
"weight": 1
},
{
"id": "odds-against:rounding",
"weight": 1
}
],
"settlement_item_field": "settlement_stratum",
"settlement_rule": "manifest-weighted arms and value; every stratum load-bearing",
"calibration": {
"planted_arm": "ainglish",
"min_gap": 0.5,
"min_recovered": null,
"rule": "absolute-gap-v1",
"ordering": "calibration-first",
"arm_exposure": "both-arms-per-reader-item",
"cells": 16
},
"difficulty": {
"annotated": false
},
"harness": "ainglish-panel/0.2.60",
"transport": {
"mistral-small3.2-24b-opaque-choice-q4_k_m@q4_k_m": {
"max_tokens": 64,
"timeout_s": 120,
"temperature": 0,
"seed": 2026090581,
"top_p": "provider-default",
"top_k": "provider-default",
"num_ctx": "provider-default",
"reasoning_effort": "provider-default"
},
"gemma3-12b-opaque-choice-q4_k_m@q4_k_m": {
"max_tokens": 64,
"timeout_s": 120,
"temperature": 0,
"seed": 2026090581,
"top_p": "provider-default",
"top_k": "provider-default",
"num_ctx": "provider-default",
"reasoning_effort": "provider-default"
}
},
"concurrency": {
"max_in_flight": 1,
"per_reader_max_in_flight": {
"mistral-small3.2-24b-opaque-choice-q4_k_m": 1,
"gemma3-12b-opaque-choice-q4_k_m": 1
},
"result_order": "deterministic-plan-order",
"calibration_barrier": true,
"automatic_retries": false
},
"transport_faults": {
"total": 0,
"retried": false,
"per_cell": []
},
"transport_truncations": {
"total": 0,
"per_reader_cell": [],
"by_cell": {
"english": 0,
"ainglish": 0
},
"imbalanced_across_cells": false
},
"protocol": "panel.py counterbalanced real arms + both-arms-per-reader-item planted-effect calibration gate",
"replicates_hash": "f270857d598a65b32d12b172773219e48e5c71950dc0dd4940f8bfddd081b4ee"
}