INDEX
Explanations
sentences or phrases with mentions of a news source (CNN) followed by a verb or action
punctuations, specifically parentheses
New Auto-Interp
Negative Logits
undergrad
-0.66
rew
-0.66
blo
-0.64
ishable
-0.64
gener
-0.62
unia
-0.62
bed
-0.62
giveaway
-0.61
ãĥĥãĥī
-0.61
repay
-0.61
POSITIVE LOGITS
âĶľ
0.79
zanne
0.78
FAM
0.71
Protective
0.71
ebus
0.70
hester
0.69
eneg
0.69
HCR
0.69
culosis
0.68
Parker
0.68
Activations Density 0.067%