Triple
T4922574
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Pillar of Severity |
E110499
|
entity |
| Predicate | bottomSephirah |
P21757
|
FINISHED |
| Object | Hod |
—
|
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: Hod | Statement: [Pillar of Severity, bottomSephirah, Hod]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: bottomSephirah Context triple: [Pillar of Severity, bottomSephirah, Hod]
-
A.
bottomSefirah
chosen
Indicates a relationship where one sefirah is identified as the lowest or final sefirah within a hierarchical structure of sefirot.
-
B.
bottomType
Indicates that the subject entity is classified as a bottom-level or most specific type within a type hierarchy.
-
C.
lowerTerminus
Indicates that one entity serves as the lower endpoint or downstream terminus of another entity, such as a route, segment, or connection.
-
D.
lowestPoint
Indicates that one entity is the point with the minimum vertical position or value relative to another entity or within a specified context.
-
E.
zenithUnder
Indicates that one entity is positioned directly below another entity’s highest or uppermost point.
- 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_69bd4413f9908190afcff44d7929cc4c |
completed | March 20, 2026, 12:56 p.m. |
| NER | Named-entity recognition | batch_69bd6ffc2ab08190992db2400562bcee |
completed | March 20, 2026, 4:04 p.m. |
| PD | Predicate disambiguation | batch_69bd6c3421588190ab08e92b9558042e |
completed | March 20, 2026, 3:48 p.m. |
Created at: March 20, 2026, 1:30 p.m.