INDEX
Explanations
sentences ending with a period
sentence-ending punctuation, particularly periods and exclamation marks
New Auto-Interp
Negative Logits
allowance
-0.83
naive
-0.72
objective
-0.72
employer
-0.71
adal
-0.69
naïve
-0.69
sympath
-0.67
minim
-0.67
agent
-0.67
uly
-0.67
POSITIVE LOGITS
Featuring
1.14
Located
1.08
Thankfully
1.05
Speaking
1.04
According
1.03
Specifically
1.03
Especially
1.02
Luckily
1.01
Apparently
0.99
Turns
0.98
Activations Density 0.551%