INDEX
Explanations
This neuron did not activate on these samples, suggesting it targets a pattern not present here (i.e., effectively inactive in this context).
New Auto-Interp
Negative Logits
ಕ್ತಿ
-0.09
ुओं
-0.09
ಕ್ತ
-0.08
/night
-0.08
glycer
-0.08
gleichen
-0.08
Fans
-0.08
Nights
-0.08
GST
-0.08
GIR
-0.08
POSITIVE LOGITS
venue
0.08
possibility
0.08
0.08
**
0.07
mynd
0.07
ollider
0.07
0.07
択
0.07
obviously
0.07
...↵
0.07
Activations Density 0.031%