INDEX

Explanations

references to a specific placeholder word indicating a general, unspecified item or concept

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

 Others

-0.71

HasForeignKey

-0.68

 surla

-0.65

 Earlier

-0.64

IsMutable

-0.63

addGap

-0.62

ufige

-0.61

TestingModule

-0.60

shund

-0.58

 يتيمه

-0.58

POSITIVE LOGITS

Any

1.73

Any

1.68

ANY

0.82

 Anytime

0.77

 Cualquier

0.71

 Qualquer

0.70

 Anything

0.69

Anyone

0.69

Anything

0.69

 Anybody

0.64

Activations Density 0.008%