INDEX

Explanations

closing punctuation or numbers

' t' follows quote

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

<unused2164>

0.48

<unused2130>

0.40

<unused2141>

0.34

<unused320>

0.32

<unused2213>

0.32

 Riemannian

0.31

<unused2084>

0.31

<unused2129>

0.31

 operates

0.30

 siano

0.30

POSITIVE LOGITS

사

0.38

ча

0.37

ſ

0.35

шки

0.34

ג

0.33

와

0.33

с

0.33

 может

0.33

ра

0.32

 στο

0.31

Activations Density 0.396%