INDEX
Explanations
phrases related to initiating or having discussions
occurrences of the word "conversation."
New Auto-Interp
Negative Logits
redit
-0.79
emale
-0.75
elin
-0.73
hee
-0.70
rule
-0.70
addon
-0.70
imposed
-0.69
azy
-0.67
evil
-0.65
uilt
-0.65
POSITIVE LOGITS
conversation
1.20
conversations
1.04
banter
0.95
Conversation
0.95
dialogue
0.88
Convers
0.88
ogue
0.83
chatter
0.80
overheard
0.80
dayName
0.78
Activations Density 0.016%