AlignmentResearch
/

probe-inference-weights

activation-probes

interpretability

Model card Files Files and versions

probe-inference-weights / qwen3.5-2b /linear /layer_7

11.1 kB

Ctrl+K

Ctrl+K

3 contributors

History: 2 commits

chrisjcundy's picture

Camera-ready probes: 9 models x {linear, mlp, efc, axial} trained on FIBSv1 @ 65dccf12

05def1c verified about 11 hours ago

config.json

128 Bytes
Add probe weights: 7 models x {linear, mlp, efc, axial} 7 days ago
model.pt

10.9 kB
xet

Camera-ready probes: 9 models x {linear, mlp, efc, axial} trained on FIBSv1 @ 65dccf12 about 11 hours ago