INDEX
Explanations
phrases where the speaker addresses or quotes someone saying "you know"
the phrase "you know," indicating conversational or informal dialogue
New Auto-Interp
Negative Logits
mage
-0.74
eta
-0.70
INAL
-0.70
ishes
-0.68
onding
-0.68
Engineers
-0.67
aval
-0.67
acs
-0.67
oreal
-0.66
oom
-0.65
POSITIVE LOGITS
sir
0.85
uh
0.84
amen
0.70
there
0.67
yeah
0.65
Lit
0.62
whoever
0.62
maybe
0.61
SPACE
0.61
knowing
0.61
Activations Density 0.105%