INDEX
Explanations
references to scientific publications and research journals
New Auto-Interp
Head Attr Weights
0:0.02
1:0.08
2:0.09
3:0.03
4:0.04
5:0.10
6:0.11
7:0.12
8:0.08
9:0.07
10:0.12
11:0.09
Negative Logits
squats
-1.24
squat
-1.09
querade
-1.06
hates
-1.05
tsky
-1.02
rises
-0.98
tatt
-0.97
yours
-0.97
awoken
-0.97
factor
-0.96
POSITIVE LOGITS
millenn
1.10
publication
1.08
Scientific
1.08
DOI
1.08
TAG
1.06
ortium
1.04
ournal
1.03
Canaver
1.03
ocamp
1.02
orthern
1.02
Activations Density 0.031%