INDEX

Explanations

dates and development information

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

 nine

-0.91

<0xA1>

-0.89

 Ninth

-0.89

九

-0.82

 Scott

-0.79

-0.78

 Nine

-0.75

 楽天

-0.75

 Initialise

-0.74

須

-0.72

POSITIVE LOGITS

इ

0.72

Александр

0.71

 ilustración

0.70

 biscuits

0.70

alignItems

0.70

 benefici

0.70

('"

0.69

 Thom

0.67

xz

0.67

0.66

Activations Density 2.231%