INDEX
Explanations
expressions of emotional states and feelings
New Auto-Interp
Negative Logits
headless
-0.61
mädchen
-0.61
rhestr
-0.59
colorir
-0.58
wicz
-0.57
Lawton
-0.56
rmızı
-0.55
uyen
-0.55
Hawkes
-0.55
برانيه
-0.55
POSITIVE LOGITS
felt
3.62
Felt
3.34
Felt
3.33
felt
3.15
feels
2.56
Feels
2.46
Feels
2.37
feel
2.37
feels
2.26
Feel
2.16
Activations Density 0.064%