INDEX

Explanations

variable names and types

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

 Cristóbal

-0.83

ӿ

-0.80

плення

-0.78

sedia

-0.76

 preocupado

-0.75

Rays

-0.75

بیر

-0.75

DOI

-0.73

Πε

-0.73

 kiện

-0.72

POSITIVE LOGITS

 negative

0.90

 length

0.88

inium

0.86

 lengths

0.86

Var

0.84

 cuánto

0.82

 variable

0.81

 Negative

0.81

 Variable

0.80

getVar

0.80

Activations Density 0.024%