INDEX
Explanations
words related to physical burning and injuries
terms related to injuries and burns
New Auto-Interp
Negative Logits
Dupl
-0.65
opp
-0.62
ack
-0.61
neau
-0.60
drop
-0.59
etsy
-0.59
drop
-0.58
trend
-0.58
zig
-0.58
exclude
-0.57
POSITIVE LOGITS
burns
3.68
bruises
1.56
wounds
1.47
boils
1.34
scars
1.26
burn
1.24
Burns
1.17
injuries
1.15
burn
1.13
burned
1.12
Activations Density 0.020%