INDEX
Explanations
phrases related to physical body parts, specifically shoulders
occurrences of the word "shoulder."
New Auto-Interp
Negative Logits
nces
-0.75
ient
-0.70
ames
-0.67
Scient
-0.63
acle
-0.63
arium
-0.62
Newman
-0.61
cult
-0.60
--------------------
-0.59
Sec
-0.59
POSITIVE LOGITS
shoulder
3.87
shoulders
2.60
elbow
1.94
knee
1.75
forearm
1.73
thigh
1.65
ankle
1.59
oulder
1.57
wrist
1.47
elbows
1.35
Activations Density 0.020%