INDEX
Explanations
references to a specific bird, possibly "Bird's Head Hagadah"
references to a specific bird or related terms
New Auto-Interp
Negative Logits
FINE
-0.78
Vanderbilt
-0.67
ilitary
-0.67
Wik
-0.66
ensions
-0.66
Uzbek
-0.65
oldemort
-0.64
anonymous
-0.62
gp
-0.62
bered
-0.61
POSITIVE LOGITS
watching
1.05
seed
1.05
bird
1.01
bath
0.96
birds
0.94
feed
0.94
walk
0.93
wat
0.90
hawk
0.90
bats
0.86
Activations Density 0.025%