INDEX
Explanations
Restaurants and food
This neuron detects the names of restaurants or eateries in the text.
New Auto-Interp
Negative Logits
अगर
-0.07
_disc
-0.07
dek
-0.07
quotations
-0.07
dalších
-0.07
-0.06
Categoria
-0.06
ridiculous
-0.06
prompts
-0.06
Wei
-0.06
POSITIVE LOGITS
......
0.07
arence
0.06
cpt
0.06
chaque
0.06
.builder
0.06
ボ
0.06
mando
0.06
("-0.06
克
0.06
อบ
0.06
Activations Density 0.027%