INDEX
Explanations
“it’s…” / “i…” / interjections
New Auto-Interp
Negative Logits
delectable
1.01
swiftly
0.89
aisément
0.83
نیز
0.82
slechts
0.78
snugly
0.73
deftly
0.73
hurriedly
0.73
Upon
0.71
столь
0.71
POSITIVE LOGITS
なんか
1.02
すごく
1.01
people
1.00
people
0.99
maybe
0.96
Maybe
0.96
наверное
0.95
خیلی
0.92
really
0.92
একটা
0.92
Activations Density 0.047%