INDEX
Explanations
names or words containing the sequence "ih"
the repeated phrase "ih" which appears frequently in the text
New Auto-Interp
Negative Logits
yards
-0.72
ãĥ¯
-0.69
Thumbnails
-0.67
Chips
-0.66
âĸ¬
-0.64
Jungle
-0.64
Blizzard
-0.63
Predator
-0.63
chips
-0.62
Bots
-0.62
POSITIVE LOGITS
ih
1.11
onen
1.09
ype
1.07
ollow
0.99
uana
0.99
oney
0.99
awk
0.95
ahah
0.94
ansen
0.93
onda
0.93
Activations Density 0.009%