INDEX
Explanations
the punctuation that marks the end of sentences or significant pauses
New Auto-Interp
Negative Logits
]")]
-1.03
$.
-0.97
")));
-0.88
".
-0.88
}}$}
-0.85
^(@)
-0.83
raiſ
-0.82
myſelf
-0.82
houſe
-0.81
itſelf
-0.81
POSITIVE LOGITS
es
0.65
v
0.63
'
0.63
b
0.63
u
0.62
e
0.61
ne
0.60
a
0.60
n
0.59
ang
0.59
Activations Density 0.766%