INDEX

Explanations

parenthesized symbols or variables

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

</b>

-2.17

他的

-1.66

</h5>

-1.48

':

-1.45

şek

-1.39

 שהוא

-1.39

</code>

-1.36

")));

-1.35

</caption>

-1.34

 sientes

-1.34

POSITIVE LOGITS

‘‘

1.43

obut

1.41

vacy

1.40

1.35

;\\

1.34

during

1.32

crats

1.30

’;

1.30

=\

1.30



1.29

Activations Density 0.068%