INDEX
Explanations
various verb forms and past participles associated with actions or states
New Auto-Interp
Head Attr Weights
0:0.07
1:0.03
2:0.11
3:0.10
4:0.08
5:0.17
6:0.03
7:0.06
8:0.06
9:0.06
10:0.12
11:0.06
Negative Logits
inav
-1.35
nih
-1.14
iga
-1.14
otto
-1.14
Sunshine
-1.14
acci
-1.14
Week
-1.13
oğ
-1.13
ggle
-1.13
ullah
-1.13
POSITIVE LOGITS
etheless
1.36
ディ
1.35
srf
1.34
!--
1.30
VIDE
1.30
comprom
1.28
android
1.26
theless
1.25
WARN
1.24
pmwiki
1.17
Activations Density 0.224%