INDEX
Explanations
occurrences of the pronoun "er" and variations of it in various descriptive contexts
New Auto-Interp
Negative Logits
vyk
-0.18
ray
-0.17
504
-0.16
vem
-0.16
_inches
-0.15
ÑĢиÑĩ
-0.15
æ§
-0.14
crest
-0.14
norm
-0.14
udent
-0.13
POSITIVE LOGITS
inner
0.23
erb
0.21
ster
0.21
Er
0.21
kan
0.20
blind
0.19
stm
0.19
hi
0.19
ford
0.18
öff
0.18
Activations Density 0.006%