INDEX

Explanations

referring to a concept

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

それとも

-1.60

};

-1.50

 zarówno

-1.46

つづく

-1.46

"[

-1.45

哎呀

-1.45

婀

-1.44

ことがある

-1.41

":

-1.41

屹

-1.40

POSITIVE LOGITS

we

1.29

 koll

1.26

 also

1.25

 universelle

1.25

guz

1.24

 gitar

1.23

woch

1.23

 nader

1.23

 gier

1.23

 atak

1.20

Activations Density 0.154%