INDEX
Explanations
instances of the verb "to be" in various forms
New Auto-Interp
Negative Logits
366
-0.16
rok
-0.15
uite
-0.15
Fist
-0.14
Suites
-0.14
irit
-0.14
owe
-0.14
airo
-0.14
386
-0.13
allee
-0.13
POSITIVE LOGITS
olas
0.16
åĨł
0.14
azor
0.14
now
0.14
ãĥ³ãĥIJ
0.14
ready
0.14
¶Į
0.13
lion
0.13
-fl
0.13
imdi
0.13
Activations Density 0.152%