INDEX
Explanations
mentions of the word "screw" in various contexts
mentions of screws and related hardware
New Auto-Interp
Negative Logits
mediate
-0.71
Nutr
-0.68
nonetheless
-0.68
ocation
-0.67
ulia
-0.67
eous
-0.66
active
-0.65
Beir
-0.64
¥µ
-0.64
aud
-0.64
POSITIVE LOGITS
screw
3.61
screws
3.11
Screw
2.69
screwed
2.23
wrench
1.39
goof
1.34
messed
1.32
fuck
1.30
fucked
1.28
messing
1.26
Activations Density 0.016%