INDEX
Explanations
Chinese names and associated actions
New Auto-Interp
Negative Logits
なか
0.61
exempl
0.58
stockfoto
0.57
ᐈ
0.55
dân
0.54
锺
0.54
romes
0.53
ঞ্চলিক
0.53
dä
0.53
髋
0.52
POSITIVE LOGITS
smiled
1.04
frowned
1.00
glanced
1.00
laughed
0.98
nodded
0.97
looked
0.97
patted
0.94
chuckled
0.93
sighed
0.92
glared
0.88
Activations Density 0.009%