INDEX
Explanations
terms related to handedness and physical manipulation
New Auto-Interp
Negative Logits
ohana
-0.16
agens
-0.15
ÐľÐŀ
-0.15
conn
-0.15
eskort
-0.15
iveau
-0.15
ÑĦÑĦ
-0.15
ÃĵN
-0.15
aget
-0.15
elsewhere
-0.15
POSITIVE LOGITS
left
0.37
Left
0.30
LEFT
0.29
left
0.29
right
0.28
Left
0.28
-left
0.28
:left
0.27
(left
0.26
direction
0.26
Activations Density 0.174%