```json { "table_name": "neal_readings", "table_type_tags": [ {"tag": "dimension", "confidence": 0.75}, {"tag": "misc", "confidence": 0.25} ], "primary_type": "dimension", "insufficient_evidence": false, "entity": { "columns": ["id"], "unique_per_row": true }, "time_axis": { "columns": ["observed_at"], "kind": "event_time" }, "reasoning": "The entity id is unique per row (Uniqueness 1.000, RPUV 1.00). Each row describes a pre-computed statistical reading (group_count, total, top3_share, top1_share) for a specific combination of source_table, group_axes, measure, and reading_kind. The observed_at timestamp has only 2 distinct values covering all 240 rows (RPUV 120.00), indicating batch computation times rather than a meaningful event axis ordering individual readings. The rows are catalog entries of computed statistics, not actions or state transitions — dimension. The composite (batch_id, source_table, group_axes, measure, reading_kind) would also be unique and represents the natural grain, but id suffices as the entity identifier." } ```