INDEX
Explanations
actions related to jumping or movement
New Auto-Interp
Negative Logits
sufficiente
-0.71
olato
-0.63
policiers
-0.63
Roh
-0.58
Britton
-0.58
Canal
-0.58
Roh
-0.56
′
-0.55
знает
-0.54
Paulson
-0.54
POSITIVE LOGITS
jump
1.83
Jump
1.82
JUMP
1.77
jumps
1.72
jump
1.67
Jump
1.63
jumping
1.55
JUMP
1.55
Leap
1.53
Jumping
1.50
Activations Density 0.088%