Gradients for Interventions and Activations for Detection: Targeted Feature Learning in Language Models
Paper • 2609.32355 • Published
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Understanding or Memorizing? A Case Study of German Definite Articles in Language Models
GRADIEND: Monosemantic Feature Learning within Neural Networks Applied to Gender Debiasing of Transformer Models
jdrechsel
authored 2
papers about 2 hours ago
jdrechsel
updated 2
datasets about 2 hours ago
jdrechsel
updated 7
datasets about 3 hours ago
aieng-lab/gradiend-key-value
Viewer • Updated • 30k • 121
aieng-lab/gradiend-ravel-neutral
Viewer • Updated • 15k • 71
aieng-lab/gradiend-ravel-language
Viewer • Updated • 660 • 68
aieng-lab/gradiend-ravel-continent
Viewer • Updated • 1.68k • 78
aieng-lab/gradiend-ravel-country
Viewer • Updated • 510 • 75
aieng-lab/gradiend-mib-ioi
Viewer • Updated • 21k • 62
aieng-lab/gradiend-repetition
Viewer • Updated • 10k • 99
jdrechsel
published 9
datasets 18 days ago
aieng-lab/gradiend-ravel-neutral
Viewer • Updated • 15k • 71
aieng-lab/gradiend-ravel-language
Viewer • Updated • 660 • 68
aieng-lab/gradiend-ravel-country
Viewer • Updated • 510 • 75
aieng-lab/gradiend-ravel-continent
Viewer • Updated • 1.68k • 78
aieng-lab/gradiend-mib-ioi
Viewer • Updated • 21k • 62
aieng-lab/gradiend-repetition
Viewer • Updated • 10k • 99
aieng-lab/gradiend-function-composition
Viewer • Updated • 30k • 161
aieng-lab/gradiend-induction
Viewer • Updated • 30k • 135
aieng-lab/gradiend-key-value
Viewer • Updated • 30k • 121