ELK (Eliciting Latent Knowledge)
GPTKB entity
Statements (14)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
AI alignment research project
|
| gptkbp:aimsTo |
extract latent knowledge from AI systems
|
| gptkbp:announced |
2021
|
| gptkbp:challenge |
eliciting true beliefs from models
|
| gptkbp:developedBy |
gptkb:Anthropic
|
| gptkbp:documentation |
gptkb:Eliciting_Latent_Knowledge_paper
|
| gptkbp:focusesOn |
AI alignment
AI interpretability |
| https://www.w3.org/2000/01/rdf-schema#label |
ELK (Eliciting Latent Knowledge)
|
| gptkbp:relatedTo |
gptkb:machine_learning
AI safety |
| gptkbp:website |
https://www.alignmentforum.org/posts/6zq4E3eK3K6eC6ZzA/eliciting-latent-knowledge
|
| gptkbp:bfsParent |
gptkb:Paul_Christiano
|
| gptkbp:bfsLayer |
6
|