INDEX
Explanations
discourse markers indicating transition or connection in sentences
New Auto-Interp
Head Attr Weights
0:0.06
1:0.03
2:0.05
3:0.13
4:0.14
5:0.06
6:0.05
7:0.03
8:0.12
9:0.15
10:0.07
11:0.03
Negative Logits
WAR
-1.39
MSN
-1.23
lees
-1.19
BuyableInstoreAndOnline
-1.17
mire
-1.16
veland
-1.16
DEM
-1.14
CV
-1.14
jab
-1.12
Prix
-1.12
POSITIVE LOGITS
quo
1.36
guessed
1.24
preempt
1.18
ende
1.16
undone
1.14
veiled
1.14
compensated
1.13
disclaim
1.11
utan
1.11
endif
1.10
Activations Density 0.008%