INDEX
Explanations
words related to art, especially when referring to creative expressions like painting, music or performance
references to art-related concepts or discussions
New Auto-Interp
Negative Logits
pora
-0.76
cumbers
-0.64
snowy
-0.61
burden
-0.61
ascus
-0.59
earthqu
-0.59
EStream
-0.59
Crosby
-0.58
dime
-0.58
Benghazi
-0.57
POSITIVE LOGITS
ifact
1.19
illery
1.13
ificial
1.04
isans
0.98
esian
0.96
ooth
0.96
icles
0.95
oon
0.94
art
0.92
ICLE
0.91
Activations Density 0.015%