INDEX
Explanations
instances of the word "about."
New Auto-Interp
Negative Logits
ردÙĩ
-0.16
)↵↵↵↵↵↵↵↵
-0.16
оÑĢоз
-0.15
posables
-0.15
VAS
-0.15
ÑĢади
-0.14
/Gate
-0.14
-pin
-0.14
lint
-0.13
boz
-0.13
POSITIVE LOGITS
-face
0.22
e
0.19
IQUE
0.19
ts
0.18
eck
0.17
-NLS
0.17
twice
0.17
to
0.17
ique
0.16
ready
0.16
Activations Density 0.071%