INDEX
Explanations
references to the letter "K," potentially indicating a focus on specific names or places associated with that letter
New Auto-Interp
Negative Logits
iyah
-0.16
iben
-0.15
ACK
-0.15
ÑĢÑĥб
-0.15
ikt
-0.15
jÃł
-0.15
ayet
-0.15
åıij
-0.14
ayas
-0.14
uang
-0.14
POSITIVE LOGITS
ritt
0.22
hand
0.20
umb
0.19
aly
0.18
irit
0.17
hus
0.17
ship
0.17
rip
0.17
hes
0.17
rish
0.16
Activations Density 0.025%