Jagged Intelligence
Overview
LLM performance is “jagged”: excels in specific tasks (e.g. coding certain patterns) but fails unexpectedly in adjacent areas. Unlike smooth human learning curves.
Key Facts
- From The Kaparthy LLM Wiki: Peaks/valleys in capabilities; expect surprises.
- Implications: Narrow specialization over general intelligence.
Related Entities
| Entity | Description |
|---|---|
| Andrej Karpathy | Coined/observed in LLM Wiki. |
Related Concepts
| Concept | Link |
|---|---|
| Vibe Coding | Leverages peaks. |
Sources/References
Connections
- Complements LLM-Wiki: Tailor ingests to peaks.
- Vs Agentic AI: Agents smooth jaggedness via tools.