INDEX
Explanations
the verb "be" in various contexts and forms
New Auto-Interp
Negative Logits
iven
-0.17
toolbox
-0.15
-toast
-0.15
_ATOMIC
-0.15
enz
-0.14
Busy
-0.14
rael
-0.14
ropolitan
-0.14
άÏĤ
-0.14
sublist
-0.13
POSITIVE LOGITS
.tel
0.18
YRO
0.16
gabe
0.15
ONTAL
0.15
hos
0.14
perc
0.14
oho
0.14
ãĤ»ãĥ³
0.14
eya
0.14
astr
0.13
Activations Density 0.008%