INDEX
Explanations
names or proper nouns containing the syllable "zh"
occurrences of specific place names or entities
New Auto-Interp
Negative Logits
staking
-0.71
loaded
-0.69
#$#$
-0.68
erc
-0.67
Kenny
-0.66
rete
-0.66
STER
-0.65
AUT
-0.63
vine
-0.62
Ready
-0.62
POSITIVE LOGITS
Zh
1.12
chens
0.95
chn
0.86
oran
0.85
nces
0.72
Tian
0.71
nikov
0.70
arnaev
0.68
py
0.68
uner
0.67
Activations Density 0.031%