INDEX
Explanations
words related to discomfort or unease
the presence of an end-of-text token, indicating the conclusion of a document or piece of writing
New Auto-Interp
Negative Logits
ball
-0.72
ãĥ£
-0.70
stones
-0.70
bing
-0.70
Hots
-0.69
Samson
-0.66
arov
-0.65
antine
-0.65
ulators
-0.65
FACE
-0.65
POSITIVE LOGITS
ebus
0.80
imate
0.75
imation
0.72
silence
0.70
schild
0.70
asonic
0.69
igma
0.68
ively
0.68
edited
0.67
Creep
0.66
Activations Density 0.064%