INDEX
Explanations
instances of the word "smile"
occurrences of the word "smile" and its variations
New Auto-Interp
Negative Logits
æ©Ł
-0.85
FER
-0.73
aer
-0.70
assic
-0.70
lay
-0.69
Administ
-0.69
raped
-0.67
Ranked
-0.67
ãĥ¯ãĥ³
-0.66
ctr
-0.65
POSITIVE LOGITS
heet
0.89
goodbye
0.86
creen
0.85
hello
0.84
smile
0.82
emot
0.78
bler
0.73
smiles
0.72
grin
0.70
ys
0.70
Activations Density 0.020%