INDEX
Negative Logits
ceries
0.76
là
0.74
include
0.73
Multiply
0.73
Include
0.73
салы
0.73
subplots
0.71
αποτε
0.70
contain
0.70
Contain
0.69
POSITIVE LOGITS
thinks
2.91
knows
2.85
wants
2.80
understands
2.64
prefers
2.64
chooses
2.63
expects
2.54
believes
2.49
refuses
2.43
perceives
2.38
Activations Density 0.161%