INDEX

Explanations

advocate for or promote

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

峮

-1.62

才可以

-1.58

点此

-1.55

でご

-1.54

鍍

-1.47

 excepcional

-1.46

 môžu

-1.45

 tournée

-1.43

 naše

-1.42

皝

-1.41

POSITIVE LOGITS

for

1.69

 fundamentals

1.55

 Resultat

1.52

但

1.51

 mengak

1.49

 menerapkan

1.43

 hond

1.41

 memeriksa

1.39

 equality

1.38

 meneg

1.38

Activations Density 0.032%