INDEX
Explanations
[entity] can/is [description]
New Auto-Interp
Negative Logits
barely
1.18
probably
0.94
mostly
0.91
largely
0.87
शायद
0.84
probably
0.84
mostly
0.83
hardly
0.83
fruitless
0.82
scarcely
0.82
POSITIVE LOGITS
っております
0.87
"!
0.79
namelijk
0.76
sometimes
0.76
Sometimes
0.75
”!
0.73
Podczas
0.71
'!
0.71
иногда
0.70
atzen
0.69
Activations Density 0.354%