INDEX
Explanations
expressions of empathy and reassurance
New Auto-Interp
Negative Logits
dal
-0.16
ainless
-0.15
oto
-0.15
ohl
-0.15
oba
-0.14
rown
-0.14
olly
-0.14
λη
-0.14
eways
-0.14
eware
-0.14
POSITIVE LOGITS
soon
0.24
Soon
0.23
Soon
0.22
eventually
0.21
.scalablytyped
0.21
soon
0.20
eventual
0.19
Eventually
0.18
sooner
0.16
WILL
0.16
Activations Density 0.207%