INDEX
Explanations
This neuron detects occurrences of the phrase “United States of America” in legal case headings.
New Auto-Interp
Negative Logits
ilingual
-0.07
Horde
-0.06
tiler
-0.06
Ruf
-0.06
�
-0.06
-trash
-0.06
’h
-0.06
.presenter
-0.06
Rahul
-0.06
Editors
-0.06
POSITIVE LOGITS
apatkan
0.07
że
0.06
dan
0.06
642
0.06
.idea
0.06
integration
0.06
Би
0.06
/templates
0.06
ंध
0.06
646
0.06
Activations Density 0.000%