INDEX

Explanations

Behind every, JUST ENTERED

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

兩

-1.88

了

-1.45

 complet

-1.31

 Juifs

-1.31

立

-1.30

 comunidade

-1.29

ükü

-1.29

往下

-1.27

 subscri

-1.27

 самой

-1.27

POSITIVE LOGITS

and

1.83

use

1.77

 while

1.47

 digunakan

1.30

 statunit

1.27

 used

1.25

ėte

1.25

↵

1.25

…！

1.23

 déficit

1.23

Activations Density 0.019%