INDEX

Explanations

forming

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

 Thou

-0.44

 gustaría

-0.44

 estés

-0.44

 utilisez

-0.43

emploi

-0.42

раздо

-0.42

قرأ

-0.42

 tér

-0.42

 Meksiku

-0.42

visning

-0.42

POSITIVE LOGITS

the

1.13

verwijspagina

0.68

 your

0.67

its

0.67

0.65

sizeCache

0.64

his

0.62

 BoxFit

0.61

 this

0.60

 BorderSide

0.58

Activations Density 0.008%