INDEX
Explanations
occurrences of the word "hence" and its contextual variations
New Auto-Interp
Negative Logits
ohn
-0.09
anca
-0.07
inya
-0.07
anie
-0.07
ignKey
-0.07
orie
-0.07
nej
-0.07
rott
-0.07
runner
-0.07
olec
-0.07
POSITIVE LOGITS
forth
0.15
forward
0.07
ìį¨
0.07
forth
0.06
ly
0.06
ful
0.06
acer
0.06
confront
0.06
hin
0.06
GBT
0.06
Activations Density 0.009%