INDEX
Explanations
phrases signaling temporal transitions and reflections
New Auto-Interp
Negative Logits
enville
-0.15
ursed
-0.15
linkplain
-0.14
oS
-0.13
aggio
-0.13
cke
-0.13
ughters
-0.13
clar
-0.13
allest
-0.13
ummies
-0.13
POSITIVE LOGITS
elage
0.16
elo
0.15
vu
0.15
ikk
0.15
318
0.14
itz
0.14
reon
0.14
ìĺ¥
0.14
UPS
0.14
.xr
0.14
Activations Density 0.032%