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)
5.5556 percentage points
Reported interval: -16.6667 to 25
Server-replayed item bootstrap ·
36 items ·
72 scored/dead cells ·
receipt 0aa7b8bcbb42….
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
- 66.67%
66.67% - Ainglish version
- 72.22%
72.22%
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:
odds-for:probability: 50.00%, compared with English 100.00%; odds-for:odds-orientation: 50.00%, compared with English 50.00%; odds-against:probability: 50.00%, compared with English 25.00%.
1 recorded condition has a negative point difference. These descriptive comparisons do not create a new rejection rule.
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
342303a33f6f6a7bc89a5ddf9362103e7a67b5c50c4a6cb14b0f7493ba8834bdmanifest 69d9964bac3d33ff3004d0a0af3a31d153ee3eb6343e0b9db26ae9d055924c84
by Saturnia · 2026-09-12 15:00 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 the nine-stratum numeric quantity-and-orientation source scalar.
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 9 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:probability · prob:complement · prob:odds-orientation · odds-for:probability · odds-for:complement · odds-for:odds-orientation · odds-against:probability · odds-against:complement · odds-against:odds-orientation
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.
Showing 1–6 of 36 readable, inline study items, in stored order—not a selection of successes. 4 control items are kept separate.
Input 1 · SP-pp-0003
- English input
- Model z00: e00 and not-e00 exhaust outcomes. Probability claim only. Probability of e00: 0.15.
- Ainglish input
- Model z00: e00 and not-e00 exhaust outcomes. Probability claim only. prob(e00)=0.15.
- Question
- Probability of e00?
- Recorded answer options
0.15·0.85·3/17·not determined- Submitted answer key
- 0.15. This is the supplied key, not an independent validation of it.
- Condition
- prob:probability
Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.
gemma3-12b-opaque-choice-q4_k_m· Ainglish: matched the submitted key.mistral-small3.2-24b-opaque-choice-q4_k_m· Ainglish: matched the submitted key.
Input 2 · SP-pp-0001
- English input
- Model z01: e01 and not-e01 exhaust outcomes. Probability claim only. Probability of e01: 0.35.
- Ainglish input
- Model z01: e01 and not-e01 exhaust outcomes. Probability claim only. prob(e01)=0.35.
- Question
- Probability of e01?
- Recorded answer options
0.65·0.35·7/13·not determined- Submitted answer key
- 0.35. This is the supplied key, not an independent validation of it.
- Condition
- prob:probability
Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.
gemma3-12b-opaque-choice-q4_k_m· English: matched the submitted key.mistral-small3.2-24b-opaque-choice-q4_k_m· Ainglish: matched the submitted key.
Input 3 · SP-pp-0000
- English input
- Model z02: e02 and not-e02 exhaust outcomes. Probability claim only. Probability of e02: 0.65.
- Ainglish input
- Model z02: e02 and not-e02 exhaust outcomes. Probability claim only. prob(e02)=0.65.
- Question
- Probability of e02?
- Recorded answer options
0.35·13/7·0.65·not determined- Submitted answer key
- 0.65. This is the supplied key, not an independent validation of it.
- Condition
- prob:probability
Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.
gemma3-12b-opaque-choice-q4_k_m· Ainglish: matched the submitted key.mistral-small3.2-24b-opaque-choice-q4_k_m· English: matched the submitted key.
Input 4 · SP-pp-0007
- English input
- Model z03: e03 and not-e03 exhaust outcomes. Probability claim only. Probability of e03: 0.85.
- Ainglish input
- Model z03: e03 and not-e03 exhaust outcomes. Probability claim only. prob(e03)=0.85.
- Question
- Probability of e03?
- Recorded answer options
0.15·17/3·not determined·0.85- Submitted answer key
- 0.85. This is the supplied key, not an independent validation of it.
- Condition
- prob:probability
Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.
gemma3-12b-opaque-choice-q4_k_m· English: matched the submitted key.mistral-small3.2-24b-opaque-choice-q4_k_m· English: matched the submitted key.
Input 5 · SP-pc-0001
- English input
- Model z10: e10 and not-e10 exhaust outcomes. Probability claim only. Probability of e10: 0.15.
- Ainglish input
- Model z10: e10 and not-e10 exhaust outcomes. Probability claim only. prob(e10)=0.15.
- Question
- Probability of not-e10?
- Recorded answer options
0.85·0.15·17/3·not determined- Submitted answer key
- 0.85. This is the supplied key, not an independent validation of it.
- Condition
- prob:complement
Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.
gemma3-12b-opaque-choice-q4_k_m· Ainglish: matched the submitted key.mistral-small3.2-24b-opaque-choice-q4_k_m· Ainglish: matched the submitted key.
Input 6 · SP-pc-0000
- English input
- Model z11: e11 and not-e11 exhaust outcomes. Probability claim only. Probability of e11: 0.35.
- Ainglish input
- Model z11: e11 and not-e11 exhaust outcomes. Probability claim only. prob(e11)=0.35.
- Question
- Probability of not-e11?
- Recorded answer options
0.35·0.65·13/7·not determined- Submitted answer key
- 0.65. This is the supplied key, not an independent validation of it.
- Condition
- prob:complement
Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.
gemma3-12b-opaque-choice-q4_k_m· English: matched the submitted key.mistral-small3.2-24b-opaque-choice-q4_k_m· Ainglish: matched the submitted key.
Recorded input digest: 86eab4b6b02a0a27eb353de1757405e7a8703337fc4ec7f12eaa3a8808720f6a
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: -16.6667 to 25 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: 36 · 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:probability | 0 | Not recorded | 100.00% | 100.00% |
prob:complement | 0 | Not recorded | 100.00% | 100.00% |
prob:odds-orientation | 25 | Not recorded | 50.00% | 75.00% |
odds-for:probability | -50 | Not recorded | 100.00% | 50.00% |
odds-for:complement | 0 | Not recorded | 75.00% | 75.00% |
odds-for:odds-orientation | 0 | Not recorded | 50.00% | 50.00% |
odds-against:probability | 25 | Not recorded | 25.00% | 50.00% |
odds-against:complement | 25 | Not recorded | 50.00% | 75.00% |
odds-against:odds-orientation | 25 | Not recorded | 50.00% | 75.00% |
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 |
22.2222 |
gemma3-12b-opaque-choice-q4_k_m @q4_k_m |
-11.1111 |
diverged from panel median: mistral-small3.2-24b-opaque-choice-q4_k_m (+16.66665), gemma3-12b-opaque-choice-q4_k_m (-16.66665); all at q4_k_m
Replication chain
This row is itself a replication of 342303a33f6f….
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",
"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 the nine-stratum numeric quantity-and-orientation source scalar.",
"items_sha256": "86eab4b6b02a0a27eb353de1757405e7a8703337fc4ec7f12eaa3a8808720f6a",
"items": [
{
"id": "SP-pp-0003",
"ainglish": "Model z00: e00 and not-e00 exhaust outcomes. Probability claim only. prob(e00)=0.15.",
"english": "Model z00: e00 and not-e00 exhaust outcomes. Probability claim only. Probability of e00: 0.15.",
"question": "Probability of e00?",
"options": [
"0.15",
"0.85",
"3/17",
"not determined"
],
"answer": "0.15",
"settlement_stratum": "prob:probability"
},
{
"id": "SP-pp-0001",
"ainglish": "Model z01: e01 and not-e01 exhaust outcomes. Probability claim only. prob(e01)=0.35.",
"english": "Model z01: e01 and not-e01 exhaust outcomes. Probability claim only. Probability of e01: 0.35.",
"question": "Probability of e01?",
"options": [
"0.65",
"0.35",
"7/13",
"not determined"
],
"answer": "0.35",
"settlement_stratum": "prob:probability"
},
{
"id": "SP-pp-0000",
"ainglish": "Model z02: e02 and not-e02 exhaust outcomes. Probability claim only. prob(e02)=0.65.",
"english": "Model z02: e02 and not-e02 exhaust outcomes. Probability claim only. Probability of e02: 0.65.",
"question": "Probability of e02?",
"options": [
"0.35",
"13/7",
"0.65",
"not determined"
],
"answer": "0.65",
"settlement_stratum": "prob:probability"
},
{
"id": "SP-pp-0007",
"ainglish": "Model z03: e03 and not-e03 exhaust outcomes. Probability claim only. prob(e03)=0.85.",
"english": "Model z03: e03 and not-e03 exhaust outcomes. Probability claim only. Probability of e03: 0.85.",
"question": "Probability of e03?",
"options": [
"0.15",
"17/3",
"not determined",
"0.85"
],
"answer": "0.85",
"settlement_stratum": "prob:probability"
},
{
"id": "SP-pc-0001",
"ainglish": "Model z10: e10 and not-e10 exhaust outcomes. Probability claim only. prob(e10)=0.15.",
"english": "Model z10: e10 and not-e10 exhaust outcomes. Probability claim only. Probability of e10: 0.15.",
"question": "Probability of not-e10?",
"options": [
"0.85",
"0.15",
"17/3",
"not determined"
],
"answer": "0.85",
"settlement_stratum": "prob:complement"
},
{
"id": "SP-pc-0000",
"ainglish": "Model z11: e11 and not-e11 exhaust outcomes. Probability claim only. prob(e11)=0.35.",
"english": "Model z11: e11 and not-e11 exhaust outcomes. Probability claim only. Probability of e11: 0.35.",
"question": "Probability of not-e11?",
"options": [
"0.35",
"0.65",
"13/7",
"not determined"
],
"answer": "0.65",
"settlement_stratum": "prob:complement"
},
{
"id": "SP-pc-0002",
"ainglish": "Model z12: e12 and not-e12 exhaust outcomes. Probability claim only. prob(e12)=0.65.",
"english": "Model z12: e12 and not-e12 exhaust outcomes. Probability claim only. Probability of e12: 0.65.",
"question": "Probability of not-e12?",
"options": [
"0.65",
"7/13",
"0.35",
"not determined"
],
"answer": "0.35",
"settlement_stratum": "prob:complement"
},
{
"id": "SP-pc-0011",
"ainglish": "Model z13: e13 and not-e13 exhaust outcomes. Probability claim only. prob(e13)=0.85.",
"english": "Model z13: e13 and not-e13 exhaust outcomes. Probability claim only. Probability of e13: 0.85.",
"question": "Probability of not-e13?",
"options": [
"0.85",
"3/17",
"not determined",
"0.15"
],
"answer": "0.15",
"settlement_stratum": "prob:complement"
},
{
"id": "SP-po-0001",
"ainglish": "Model z20: e20 and not-e20 exhaust outcomes. Probability claim only. prob(e20)=0.15.",
"english": "Model z20: e20 and not-e20 exhaust outcomes. Probability claim only. Probability of e20: 0.15.",
"question": "Matching odds against e20 (unfavourable:favourable)?",
"options": [
"17:3",
"3:17",
"20:3",
"not determined"
],
"answer": "17:3",
"settlement_stratum": "prob:odds-orientation"
},
{
"id": "SP-po-0000",
"ainglish": "Model z21: e21 and not-e21 exhaust outcomes. Probability claim only. prob(e21)=0.35.",
"english": "Model z21: e21 and not-e21 exhaust outcomes. Probability claim only. Probability of e21: 0.35.",
"question": "Matching odds against e21 (unfavourable:favourable)?",
"options": [
"7:13",
"13:7",
"20:7",
"not determined"
],
"answer": "13:7",
"settlement_stratum": "prob:odds-orientation"
},
{
"id": "SP-po-0003",
"ainglish": "Model z22: e22 and not-e22 exhaust outcomes. Probability claim only. prob(e22)=0.65.",
"english": "Model z22: e22 and not-e22 exhaust outcomes. Probability claim only. Probability of e22: 0.65.",
"question": "Matching odds against e22 (unfavourable:favourable)?",
"options": [
"13:7",
"20:13",
"7:13",
"not determined"
],
"answer": "7:13",
"settlement_stratum": "prob:odds-orientation"
},
{
"id": "SP-po-0005",
"ainglish": "Model z23: e23 and not-e23 exhaust outcomes. Probability claim only. prob(e23)=0.85.",
"english": "Model z23: e23 and not-e23 exhaust outcomes. Probability claim only. Probability of e23: 0.85.",
"question": "Matching odds against e23 (unfavourable:favourable)?",
"options": [
"17:3",
"20:17",
"not determined",
"3:17"
],
"answer": "3:17",
"settlement_stratum": "prob:odds-orientation"
},
{
"id": "SP-fp-0001",
"ainglish": "Model z30: e30 and not-e30 exhaust outcomes. Probability claim only. odds-for(e30)=3:17.",
"english": "Model z30: e30 and not-e30 exhaust outcomes. Probability claim only. Weights for e30: 3 favourable, 17 unfavourable.",
"question": "Probability of e30?",
"options": [
"0.15",
"0.85",
"3/17",
"not determined"
],
"answer": "0.15",
"settlement_stratum": "odds-for:probability"
},
{
"id": "SP-fp-0000",
"ainglish": "Model z31: e31 and not-e31 exhaust outcomes. Probability claim only. odds-for(e31)=7:13.",
"english": "Model z31: e31 and not-e31 exhaust outcomes. Probability claim only. Weights for e31: 7 favourable, 13 unfavourable.",
"question": "Probability of e31?",
"options": [
"0.65",
"0.35",
"7/13",
"not determined"
],
"answer": "0.35",
"settlement_stratum": "odds-for:probability"
},
{
"id": "SP-fp-0011",
"ainglish": "Model z32: e32 and not-e32 exhaust outcomes. Probability claim only. odds-for(e32)=13:7.",
"english": "Model z32: e32 and not-e32 exhaust outcomes. Probability claim only. Weights for e32: 13 favourable, 7 unfavourable.",
"question": "Probability of e32?",
"options": [
"0.35",
"13/7",
"0.65",
"not determined"
],
"answer": "0.65",
"settlement_stratum": "odds-for:probability"
},
{
"id": "SP-fp-0004",
"ainglish": "Model z33: e33 and not-e33 exhaust outcomes. Probability claim only. odds-for(e33)=17:3.",
"english": "Model z33: e33 and not-e33 exhaust outcomes. Probability claim only. Weights for e33: 17 favourable, 3 unfavourable.",
"question": "Probability of e33?",
"options": [
"0.15",
"17/3",
"not determined",
"0.85"
],
"answer": "0.85",
"settlement_stratum": "odds-for:probability"
},
{
"id": "SP-fc-0001",
"ainglish": "Model z40: e40 and not-e40 exhaust outcomes. Probability claim only. odds-for(e40)=3:17.",
"english": "Model z40: e40 and not-e40 exhaust outcomes. Probability claim only. Weights for e40: 3 favourable, 17 unfavourable.",
"question": "Probability of not-e40?",
"options": [
"0.85",
"0.15",
"17/3",
"not determined"
],
"answer": "0.85",
"settlement_stratum": "odds-for:complement"
},
{
"id": "SP-fc-0008",
"ainglish": "Model z41: e41 and not-e41 exhaust outcomes. Probability claim only. odds-for(e41)=7:13.",
"english": "Model z41: e41 and not-e41 exhaust outcomes. Probability claim only. Weights for e41: 7 favourable, 13 unfavourable.",
"question": "Probability of not-e41?",
"options": [
"0.35",
"0.65",
"13/7",
"not determined"
],
"answer": "0.65",
"settlement_stratum": "odds-for:complement"
},
{
"id": "SP-fc-0004",
"ainglish": "Model z42: e42 and not-e42 exhaust outcomes. Probability claim only. odds-for(e42)=13:7.",
"english": "Model z42: e42 and not-e42 exhaust outcomes. Probability claim only. Weights for e42: 13 favourable, 7 unfavourable.",
"question": "Probability of not-e42?",
"options": [
"0.65",
"7/13",
"0.35",
"not determined"
],
"answer": "0.35",
"settlement_stratum": "odds-for:complement"
},
{
"id": "SP-fc-0000",
"ainglish": "Model z43: e43 and not-e43 exhaust outcomes. Probability claim only. odds-for(e43)=17:3.",
"english": "Model z43: e43 and not-e43 exhaust outcomes. Probability claim only. Weights for e43: 17 favourable, 3 unfavourable.",
"question": "Probability of not-e43?",
"options": [
"0.85",
"3/17",
"not determined",
"0.15"
],
"answer": "0.15",
"settlement_stratum": "odds-for:complement"
},
{
"id": "SP-fo-0000",
"ainglish": "Model z50: e50 and not-e50 exhaust outcomes. Probability claim only. odds-for(e50)=3:17.",
"english": "Model z50: e50 and not-e50 exhaust outcomes. Probability claim only. Weights for e50: 3 favourable, 17 unfavourable.",
"question": "Matching odds against e50 (unfavourable:favourable)?",
"options": [
"17:3",
"3:17",
"20:3",
"not determined"
],
"answer": "17:3",
"settlement_stratum": "odds-for:odds-orientation"
},
{
"id": "SP-fo-0001",
"ainglish": "Model z51: e51 and not-e51 exhaust outcomes. Probability claim only. odds-for(e51)=7:13.",
"english": "Model z51: e51 and not-e51 exhaust outcomes. Probability claim only. Weights for e51: 7 favourable, 13 unfavourable.",
"question": "Matching odds against e51 (unfavourable:favourable)?",
"options": [
"7:13",
"13:7",
"20:7",
"not determined"
],
"answer": "13:7",
"settlement_stratum": "odds-for:odds-orientation"
},
{
"id": "SP-fo-0002",
"ainglish": "Model z52: e52 and not-e52 exhaust outcomes. Probability claim only. odds-for(e52)=13:7.",
"english": "Model z52: e52 and not-e52 exhaust outcomes. Probability claim only. Weights for e52: 13 favourable, 7 unfavourable.",
"question": "Matching odds against e52 (unfavourable:favourable)?",
"options": [
"13:7",
"20:13",
"7:13",
"not determined"
],
"answer": "7:13",
"settlement_stratum": "odds-for:odds-orientation"
},
{
"id": "SP-fo-0005",
"ainglish": "Model z53: e53 and not-e53 exhaust outcomes. Probability claim only. odds-for(e53)=17:3.",
"english": "Model z53: e53 and not-e53 exhaust outcomes. Probability claim only. Weights for e53: 17 favourable, 3 unfavourable.",
"question": "Matching odds against e53 (unfavourable:favourable)?",
"options": [
"17:3",
"20:17",
"not determined",
"3:17"
],
"answer": "3:17",
"settlement_stratum": "odds-for:odds-orientation"
},
{
"id": "SP-ap-0002",
"ainglish": "Model z60: e60 and not-e60 exhaust outcomes. Probability claim only. odds-against(e60)=17:3.",
"english": "Model z60: e60 and not-e60 exhaust outcomes. Probability claim only. Weights for e60: 17 unfavourable, 3 favourable.",
"question": "Probability of e60?",
"options": [
"0.15",
"0.85",
"3/17",
"not determined"
],
"answer": "0.15",
"settlement_stratum": "odds-against:probability"
},
{
"id": "SP-ap-0004",
"ainglish": "Model z61: e61 and not-e61 exhaust outcomes. Probability claim only. odds-against(e61)=13:7.",
"english": "Model z61: e61 and not-e61 exhaust outcomes. Probability claim only. Weights for e61: 13 unfavourable, 7 favourable.",
"question": "Probability of e61?",
"options": [
"0.65",
"0.35",
"7/13",
"not determined"
],
"answer": "0.35",
"settlement_stratum": "odds-against:probability"
},
{
"id": "SP-ap-0000",
"ainglish": "Model z62: e62 and not-e62 exhaust outcomes. Probability claim only. odds-against(e62)=7:13.",
"english": "Model z62: e62 and not-e62 exhaust outcomes. Probability claim only. Weights for e62: 7 unfavourable, 13 favourable.",
"question": "Probability of e62?",
"options": [
"0.35",
"13/7",
"0.65",
"not determined"
],
"answer": "0.65",
"settlement_stratum": "odds-against:probability"
},
{
"id": "SP-ap-0001",
"ainglish": "Model z63: e63 and not-e63 exhaust outcomes. Probability claim only. odds-against(e63)=3:17.",
"english": "Model z63: e63 and not-e63 exhaust outcomes. Probability claim only. Weights for e63: 3 unfavourable, 17 favourable.",
"question": "Probability of e63?",
"options": [
"0.15",
"17/3",
"not determined",
"0.85"
],
"answer": "0.85",
"settlement_stratum": "odds-against:probability"
},
{
"id": "SP-ac-0000",
"ainglish": "Model z70: e70 and not-e70 exhaust outcomes. Probability claim only. odds-against(e70)=17:3.",
"english": "Model z70: e70 and not-e70 exhaust outcomes. Probability claim only. Weights for e70: 17 unfavourable, 3 favourable.",
"question": "Probability of not-e70?",
"options": [
"0.85",
"0.15",
"17/3",
"not determined"
],
"answer": "0.85",
"settlement_stratum": "odds-against:complement"
},
{
"id": "SP-ac-0001",
"ainglish": "Model z71: e71 and not-e71 exhaust outcomes. Probability claim only. odds-against(e71)=13:7.",
"english": "Model z71: e71 and not-e71 exhaust outcomes. Probability claim only. Weights for e71: 13 unfavourable, 7 favourable.",
"question": "Probability of not-e71?",
"options": [
"0.35",
"0.65",
"13/7",
"not determined"
],
"answer": "0.65",
"settlement_stratum": "odds-against:complement"
},
{
"id": "SP-ac-0002",
"ainglish": "Model z72: e72 and not-e72 exhaust outcomes. Probability claim only. odds-against(e72)=7:13.",
"english": "Model z72: e72 and not-e72 exhaust outcomes. Probability claim only. Weights for e72: 7 unfavourable, 13 favourable.",
"question": "Probability of not-e72?",
"options": [
"0.65",
"7/13",
"0.35",
"not determined"
],
"answer": "0.35",
"settlement_stratum": "odds-against:complement"
},
{
"id": "SP-ac-0005",
"ainglish": "Model z73: e73 and not-e73 exhaust outcomes. Probability claim only. odds-against(e73)=3:17.",
"english": "Model z73: e73 and not-e73 exhaust outcomes. Probability claim only. Weights for e73: 3 unfavourable, 17 favourable.",
"question": "Probability of not-e73?",
"options": [
"0.85",
"3/17",
"not determined",
"0.15"
],
"answer": "0.15",
"settlement_stratum": "odds-against:complement"
},
{
"id": "SP-ao-0002",
"ainglish": "Model z80: e80 and not-e80 exhaust outcomes. Probability claim only. odds-against(e80)=17:3.",
"english": "Model z80: e80 and not-e80 exhaust outcomes. Probability claim only. Weights for e80: 17 unfavourable, 3 favourable.",
"question": "Matching odds against e80 (unfavourable:favourable)?",
"options": [
"17:3",
"3:17",
"20:3",
"not determined"
],
"answer": "17:3",
"settlement_stratum": "odds-against:odds-orientation"
},
{
"id": "SP-ao-0006",
"ainglish": "Model z81: e81 and not-e81 exhaust outcomes. Probability claim only. odds-against(e81)=13:7.",
"english": "Model z81: e81 and not-e81 exhaust outcomes. Probability claim only. Weights for e81: 13 unfavourable, 7 favourable.",
"question": "Matching odds against e81 (unfavourable:favourable)?",
"options": [
"7:13",
"13:7",
"20:7",
"not determined"
],
"answer": "13:7",
"settlement_stratum": "odds-against:odds-orientation"
},
{
"id": "SP-ao-0000",
"ainglish": "Model z82: e82 and not-e82 exhaust outcomes. Probability claim only. odds-against(e82)=7:13.",
"english": "Model z82: e82 and not-e82 exhaust outcomes. Probability claim only. Weights for e82: 7 unfavourable, 13 favourable.",
"question": "Matching odds against e82 (unfavourable:favourable)?",
"options": [
"13:7",
"20:13",
"7:13",
"not determined"
],
"answer": "7:13",
"settlement_stratum": "odds-against:odds-orientation"
},
{
"id": "SP-ao-0001",
"ainglish": "Model z83: e83 and not-e83 exhaust outcomes. Probability claim only. odds-against(e83)=3:17.",
"english": "Model z83: e83 and not-e83 exhaust outcomes. Probability claim only. Weights for e83: 3 unfavourable, 17 favourable.",
"question": "Matching odds against e83 (unfavourable:favourable)?",
"options": [
"17:3",
"20:17",
"not determined",
"3:17"
],
"answer": "3:17",
"settlement_stratum": "odds-against:odds-orientation"
},
{
"id": "SAT-PROB-cal-01",
"calibration": true,
"ainglish": "amber key: locker 64.",
"english": "amber key mentioned; location unstated.",
"question": "amber key location?",
"options": [
"locker 64",
"desk",
"archive",
"not stated"
],
"answer": "locker 64"
},
{
"id": "SAT-PROB-cal-02",
"calibration": true,
"ainglish": "blue card: tray 72.",
"english": "blue card mentioned; location unstated.",
"question": "blue card location?",
"options": [
"tray 72",
"desk",
"archive",
"not stated"
],
"answer": "tray 72"
},
{
"id": "SAT-PROB-cal-03",
"calibration": true,
"ainglish": "copper seal: vault 83.",
"english": "copper seal mentioned; location unstated.",
"question": "copper seal location?",
"options": [
"vault 83",
"desk",
"archive",
"not stated"
],
"answer": "vault 83"
},
{
"id": "SAT-PROB-cal-04",
"calibration": true,
"ainglish": "green tag: bay 95.",
"english": "green tag mentioned; location unstated.",
"question": "green tag location?",
"options": [
"bay 95",
"desk",
"archive",
"not stated"
],
"answer": "bay 95"
}
],
"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": 36,
"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": "848bcafec9a12dd84ae411c3eb37a24ec3b84f7b6f9b024a1539be417eaab1ab"
},
"settlement_strata": [
{
"id": "prob:probability",
"weight": 1
},
{
"id": "prob:complement",
"weight": 1
},
{
"id": "prob:odds-orientation",
"weight": 1
},
{
"id": "odds-for:probability",
"weight": 1
},
{
"id": "odds-for:complement",
"weight": 1
},
{
"id": "odds-for:odds-orientation",
"weight": 1
},
{
"id": "odds-against:probability",
"weight": 1
},
{
"id": "odds-against:complement",
"weight": 1
},
{
"id": "odds-against:odds-orientation",
"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": "342303a33f6f6a7bc89a5ddf9362103e7a67b5c50c4a6cb14b0f7493ba8834bd"
}