INDEX
Explanations
important words across languages
New Auto-Interp
Negative Logits
ूरती
0.44
ぅ
0.41
⇰
0.40
lanelet
0.40
運輸
0.39
swift
0.38
ococ
0.38
駕
0.37
рома
0.37
ゅ
0.37
POSITIVE LOGITS
important
0.47
penting
0.44
важных
0.43
tärke
0.43
belangrijk
0.42
முக்கியமான
0.42
multib
0.42
important
0.41
مهم
0.41
raphics
0.40
Activations Density 0.003%