INDEX

Explanations

technical components or criteria

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

為に

-1.72

-1.38

 turut

-1.33

ちな

-1.31

ようです

-1.27

excellents

-1.26

ところに

-1.25

 this

-1.25

 refroidissement

-1.23

and

-1.23

POSITIVE LOGITS

を

1.89

が

1.63

in

1.63

です

1.55

は

1.53

1.47

1.45

の

1.45

1.39

_"

1.38

Activations Density 0.548%