INDEX
Explanations
instances of strong emotional reactions expressed through punctuation and pauses
New Auto-Interp
Negative Logits
neys
-0.16
alement
-0.16
eps
-0.16
.sponge
-0.15
ijing
-0.15
åĶĩ
-0.15
_rwlock
-0.14
romise
-0.14
екÑĤоÑĢ
-0.14
acyj
-0.14
POSITIVE LOGITS
ply
0.18
otas
0.16
712
0.15
809
0.14
Gonzalez
0.13
Lif
0.13
ADDE
0.13
outer
0.13
ignon
0.13
lif
0.13
Activations Density 0.165%