A collection of model organisms of sandbagging, that is, deliberate underperformance on a task. The task is code vulnerability detection.
Lennart Finke
lennart-finke
AI & ML interests
AI Safety
Recent Activity
updated a collection 3 days ago
sandbagging updated a collection 3 days ago
sandbagging updated a collection 3 days ago
sandbaggingOrganizations
models 8
lennart-finke/SD
36B • Updated • 11
lennart-finke/Reinforcement-Learning
36B • Updated • 12
lennart-finke/Expert-Iteration
36B • Updated • 10
lennart-finke/SDF
36B • Updated • 9
lennart-finke/Password-locked
36B • Updated • 9
lennart-finke/KTO-SFT
36B • Updated • 14
lennart-finke/Strong-TAR
35B • Updated • 15
lennart-finke/Classifier
Updated