INDEX
Explanations
reactions and responses
The neuron activates on discussion of realistic reactions and responses in the narrative—i.e. words describing “realistic reactions of the environment and other characters.”
New Auto-Interp
Negative Logits
Grant
-0.06
', ↵
-0.06
ürk
-0.06
alumni
-0.06
уля
-0.06
ーデ
-0.06
tourist
-0.06
Verde
-0.06
Ye
-0.06
carriage
-0.06
POSITIVE LOGITS
Проф
0.06
/Test
0.06
перер
0.06
052
0.06
Test
0.06
violate
0.05
modifiers
0.05
-shift
0.05
σταση
0.05
curb
0.05
Activations Density 0.004%