INDEX
Explanations
verbs describing forceful or physical action
instances of the word "dragged" and its variations in context
New Auto-Interp
Negative Logits
Correct
-0.67
uate
-0.67
ucc
-0.66
Kit
-0.65
Effective
-0.64
Pere
-0.63
Bay
-0.62
Mahar
-0.62
Pent
-0.62
ible
-0.61
POSITIVE LOGITS
dragged
3.38
dragging
2.38
drag
2.11
Drag
1.66
hauled
1.61
Drag
1.51
DRAG
1.33
crawled
1.24
sucked
1.24
shoved
1.20
Activations Density 0.014%