INDEX
Explanations
words related to body parts, specifically legs
occurrences of the word "leg"
New Auto-Interp
Negative Logits
hower
-0.73
aukee
-0.72
Tycoon
-0.70
pload
-0.67
anyahu
-0.66
ettings
-0.66
999
-0.63
ongyang
-0.62
vacancy
-0.61
cffff
-0.60
POSITIVE LOGITS
acies
1.18
isl
0.86
amput
0.86
chair
0.85
leg
0.83
locks
0.83
iev
0.83
puter
0.82
itimate
0.82
guards
0.79
Activations Density 0.013%