INDEX
Explanations
The neuron is triggered by educational context words—especially mentions of students, campuses, universities or colleges.
New Auto-Interp
Negative Logits
joke
-0.07
(ap
-0.06
_text
-0.06
.Man
-0.06
Materials
-0.06
arious
-0.06
_START
-0.06
Muslims
-0.06
orf
-0.06
Problem
-0.06
POSITIVE LOGITS
.transfer
0.07
_WORDS
0.06
Tiếng
0.06
.links
0.06
เกม
0.06
yürüy
0.06
vital
0.06
益
0.06
rencontre
0.06
:;"
0.06
Activations Density 0.019%