INDEX
Explanations
instances of the word "rather" in various forms
New Auto-Interp
Negative Logits
èle
-0.17
moreover
-0.16
ulumi
-0.16
swer
-0.16
OMET
-0.16
otta
-0.15
vie
-0.15
лив
-0.15
ys
-0.14
ateg
-0.14
POSITIVE LOGITS
than
0.49
than
0.38
Than
0.35
-than
0.33
Than
0.32
THAN
0.32
_than
0.32
než
0.30
än
0.28
_THAN
0.26
Activations Density 0.015%