INDEX
Explanations
The neuron activates on occurrences of “Tibet” and its derivatives (e.g. “Tibetan,” “Tibetans,” “Tibetan Plateau”).
New Auto-Interp
Negative Logits
ون
-0.07
ONS
-0.06
TTY
-0.06
PROVID
-0.06
Serum
-0.06
_IRQ
-0.06
USH
-0.06
enerj
-0.06
Elm
-0.06
aştır
-0.06
POSITIVE LOGITS
Tibet
0.10
Tibetan
0.10
Nepal
0.07
neben
0.07
错
0.06
inflatable
0.06
fuss
0.06
้อน
0.06
back
0.06
\")
0.06
Activations Density 0.001%