INDEX

Explanations

numbers followed by 3 or .3

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esen

-0.09

 quint

-0.09

Ð¾Ð±ÑĢÐ°Ð

-0.09

kot

-0.08

_RGBA

-0.08

uale

-0.08

eus

-0.08

/logo

-0.08

 Caul

-0.08

 Leer

-0.08

POSITIVE LOGITS

0.43

ĥ

0.39

 third

0.34

Â³

0.34

 Third

0.33

III

0.32

ï¼ĵ

0.32

 three

0.32

 THIRD

0.31

-third

0.30

Activations Density 0.857%