INDEX
Explanations
phrases related to sitting or lying down on furniture pieces like couches or sofas
references to couches and sofas
New Auto-Interp
Negative Logits
utenberg
-0.73
abad
-0.66
hypers
-0.65
uality
-0.65
Resistance
-0.64
iyah
-0.64
iferation
-0.63
letters
-0.63
phases
-0.61
nesses
-0.61
POSITIVE LOGITS
cush
1.21
washer
1.12
chair
1.04
cushion
1.04
couch
1.02
tops
0.99
sofa
0.98
potato
0.91
chairs
0.90
stairs
0.88
Activations Density 0.005%