INDEX
Explanations
repetitive phrases such as "again and again" or "more and more"
repetitive phrases and expressions indicating recurrence or intensity
New Auto-Interp
Negative Logits
Joy
-0.76
Nap
-0.72
asc
-0.72
ICO
-0.70
IV
-0.67
Disk
-0.66
Mor
-0.66
gar
-0.65
assis
-0.64
577
-0.63
POSITIVE LOGITS
alike
0.77
stray
0.70
editions
0.64
bully
0.63
hybrid
0.63
across
0.63
scrimmage
0.62
phosphorus
0.61
pace
0.61
igree
0.61
Activations Density 0.091%