INDEX
Explanations
medical research
The neuron detects mentions of specific chemical or drug names in the text.
New Auto-Interp
Negative Logits
악
-0.07
wright
-0.07
Deleting
-0.07
_with
-0.06
pond
-0.06
Prop
-0.06
uls
-0.06
astr
-0.06
'),'
-0.06
especially
-0.06
POSITIVE LOGITS
昭和
0.07
أع
0.06
Known
0.06
.U
0.06
ические
0.06
ульта
0.06
ار
0.06
oliday
0.06
只
0.06
deployed
0.06
Activations Density 0.049%