INDEX

Explanations

the tokens "mar", "MAR" and related phrases

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

indeer

-0.66

 Perseus

-0.64

 Gunther

-0.63

 méta

-0.60

work

-0.60

 munt

-0.59

therford

-0.58

htë

-0.58

pidou

-0.58

 Thakur

-0.57

POSITIVE LOGITS

Mar

2.95

MAR

2.81

mar

2.78

Mar

2.78

MAR

2.56

mar

2.55

 Мар

1.86

 мар

1.82

Мар

1.78

 Maru

1.77

Activations Density 0.401%