INDEX
Explanations
phrases indicating a collective group or general consensus
references to a collective group of people, particularly "everyone" and "everybody."
New Auto-Interp
Negative Logits
tnc
-0.74
pose
-0.72
éŃĶ
-0.71
iger
-0.65
slaught
-0.64
æĪ¦
-0.63
sole
-0.62
aye
-0.61
éļ
-0.60
é¾įå¥ij士
-0.60
POSITIVE LOGITS
else
1.58
knows
1.33
agrees
1.32
hates
1.23
loves
1.19
remembers
1.14
Else
1.12
wants
1.09
assumes
1.08
recognizes
1.06
Activations Density 0.059%