INDEX
Explanations
exclamations of surprise or realization
expressions of surprise or emphasis, particularly those that begin with "Oh."
New Auto-Interp
Negative Logits
perature
-0.72
-+-+
-0.71
":[{"-0.63
IUM
-0.62
iership
-0.61
Luxem
-0.60
ciplinary
-0.59
Construct
-0.58
esthetic
-0.57
assembled
-0.56
POSITIVE LOGITS
hhhh
1.03
dear
0.95
hhh
0.94
anian
0.92
yeah
0.89
hh
0.88
oho
0.83
yea
0.82
oy
0.80
yeah
0.80
Activations Density 0.016%