INDEX
Explanations
This neuron activates on occurrences of the word “books,” specifically when listing or recommending books.
New Auto-Interp
Negative Logits
Brew
-0.07
_survey
-0.07
kontro
-0.07
Passed
-0.07
่านมา
-0.07
Subject
-0.07
worst
-0.07
Sector
-0.07
.Before
-0.07
Overs
-0.07
POSITIVE LOGITS
Spain
0.07
book
0.07
гра
0.06
rounded
0.06
(png
0.06
btn
0.06
Books
0.06
ιλ
0.06
Spain
0.06
ksi
0.06
Activations Density 0.025%