Knowledge Graph
Graph explorer
Search resolved entities, walk their edges, and read the sentence behind every one.
Entities25 of 207,987
- large language modeltechnology1,047 mentions
- robotic manipulationmarket829 mentions
- artificial intelligencetechnology739 mentions
- reinforcement learningtechnology491 mentions
- density-functional theorytechnology386 mentions
- Transformertechnology372 mentions
- agentic AItechnology353 mentions
- vision-language-action modeltechnology352 mentions
- U.S. Food and Drug Administrationorganization320 mentions
- Vision-Language Modeltechnology320 mentions
- Microsoftorganization291 mentions
- Googleorganization276 mentions
- NVIDIAorganization264 mentions
- Model Context Protocoltechnology232 mentions
- machine learningtechnology204 mentions
- MarketsandMarketsorganization199 mentions
- Claudeproduct192 mentions
- AI agenttechnology188 mentions
- ChatGPTproduct178 mentions
- Anthropicorganization177 mentions
- integrated sensing and communicationtechnology176 mentions
- Amazon Web Services, Inc.organization174 mentions
- OpenAIorganization172 mentions
- diffusion modeltechnology169 mentions
- IBMorganization163 mentions
Entity
linear probe
technology16 mentionslast seen Nov 9, 2026
A current method for measuring reliance on concepts by asking whether each concept in isolation is decodable from a layer.
Relationships 9
- linear probeapplies tocompositional generalisation91% confidence1 source
- linear probeapplies tohallucination detection in language models90% confidence1 source
- linear probeimplementsUnderwater Acoustic Target Recognition88% confidence1 source
- conditional interventiondepends onlinear probe88% confidence1 source
- linear probedepends onpretrained audio embedding87% confidence1 source
- SYNAPSEimplementslinear probe85% confidence1 source
- vision-language-action modeldepends onlinear probe85% confidence1 source
- linear probeapplies totabular data prediction82% confidence1 source
- linear probeapplies toneural network interpretability70% confidence1 source
Relationship
linear probeapplies tohallucination detection in language models
90% confidence1 supporting mentions
Either name opens that entity’s relationships.
Sources 1