INDEX

Explanations

Negative situations

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

 Himo

-0.82

 doom

-0.75

 pretence

-0.75

 misfortunes

-0.74

 wickedness

-0.73

 miseries

-0.73

 mercies

-0.72

 woes

-0.71

 Loser

-0.70

 loveliness

-0.70

POSITIVE LOGITS

iness

0.74

ings

0.65

iest

0.63

webElementXpaths

0.56

 that

0.55

 prior

0.54

hips

0.54

ier

0.52

0.51

fulness

0.49

Activations Density 0.054%