Triple
T3377627
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Eye of Harmony |
E71103
|
entity |
| Predicate | dangerTo |
P16824
|
FINISHED |
| Object | the universe if uncontrolled |
—
|
LITERAL FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: the universe if uncontrolled | Statement: [Eye of Harmony, dangerTo, the universe if uncontrolled]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: dangerTo Context triple: [Eye of Harmony, dangerTo, the universe if uncontrolled]
-
A.
threat
Indicates a relationship where one entity expresses or poses potential harm, danger, or negative consequences toward another entity.
-
B.
endangerment
chosen
Indicates a relationship in which one entity exposes another to potential harm, risk, or danger.
-
C.
threatCategory
Indicates the classification of a threat according to its type, severity, or nature within a defined risk or security framework.
-
D.
toxicTo
Indicates that one entity causes harm, poisoning, or adverse effects to another when exposed or applied.
-
E.
inDangerSince
Indicates that an entity has been in a state of danger continuously starting from a specified point in time.
- F. None of above.
Provenance (3 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69ad85a7f80c8190a05e43013f298942 |
completed | March 8, 2026, 2:20 p.m. |
| NER | Named-entity recognition | batch_69adb2e8d1988190b6fb6c4c5502f25f |
completed | March 8, 2026, 5:33 p.m. |
| PD | Predicate disambiguation | batch_69ada433059881908e46f38cc5f40a32 |
completed | March 8, 2026, 4:30 p.m. |
Created at: March 8, 2026, 3:13 p.m.