INDEX

Explanations

hard followed by space

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

 בְּ

-0.80

 tatlı

-0.77

途

-0.76

Przypisy

-0.75

 horses

-0.71

INDA

-0.70

作品

-0.69

psie

-0.69

noticed

-0.68

fs

-0.68

POSITIVE LOGITS

 wearing

1.02

wares

0.84

wearing

0.82

dow

0.81

 hábit

0.78

nesses

0.77

coat

0.75

Wearing

0.75

 wears

0.75

 Wearing

0.74

Activations Density 0.024%