community
active
leiden_hybrid_concepts
label: haiku
community:leiden_hybrid_concepts-run4-c0-c0-c1Emergence through distributed attention and uncertainty
Explores how complex phenomena arise from non-linear interactions across distributed systems, emphasizing productive not-knowing and implicit learning mechanisms.
4 members. Each node is clickable.
Loading graph…
Drawn from 4 sources
The papers/notes whose extracted claims & findings make up this cluster.
- 2026-05-09_briefing_for_ozero.md1 member
- Paper Summary: Interpreting Language Model Parameters1 member
- unfold-chat-catalog.md1 member
- Emergence and Causality in Complex Systems: A Survey on Causal Emergence and Related Quantitative Studies1 member
Bridges (3)
Other communities that share members with this one — cross-cutting threads or papers that sit at the seam between two themes.
Claims (3)
- Incorporating machine learning provides objective standards that help mitigate subjectivity in emergence identification.Authors argue ML optimizers act as objective observers.
- Insight cascades and implicit learning require balance between directed attention and openness.
- Not-knowing, silence, incompleteness, and non-defensiveness function as positive traits, not deficits.
Findings (1)
- Identification of algorithms implemented in attention layers, distributed across attention headsVPD successfully recovered interpretable attention algorithms (previous-token behavior, syntax-boundary routing) in weight space without requiring manual decomposition across heads.