INDEX

Explanations

the word "higher"

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

-0.79

åg

-0.77

 Parton

-0.72

KAL

-0.70

</i>

-0.66

alah

-0.63

Кол

-0.62

ToolStrip

-0.62

Lars

-0.61

новниш

-0.60

POSITIVE LOGITS

complexContent

1.00

weile

0.88

 Hiller

0.84

 Mundi

0.81

moveToNext

0.79

Посилання

0.77

 okaza

0.75

 Humb

0.73

 Thistle

0.73

Demografie

0.71

Activations Density 0.022%