INDEX

Explanations

shows how something represents another

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

↵↵

-1.77

to

-1.59

 Anything

-1.55

馘

-1.47

But

-1.42

стр

-1.41

ernos

-1.38

呖

-1.37

та

-1.36

>}

-1.36

POSITIVE LOGITS

 with

1.56

 reflects

1.55

 recibieron

1.53

 Οι

1.45

 käytet

1.42

lahir

1.39

Puissance

1.39

 übri

1.38

Rebs

1.36

Karakteristik

1.36

Activations Density 0.024%