INDEX

Explanations

image 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

bertura

-0.72

 něco

-0.72

Podemos

-0.72

HEC

-0.70

uyang

-0.70

Edo

-0.70

 Frü

-0.70

Sinopse

-0.69

łosi

-0.68

Abuja

-0.68

POSITIVE LOGITS

 image

3.23

 images

2.98

 pictures

2.14

image

2.08

 picture

2.06

 photos

2.03

 photograph

1.99

 photographs

1.99

img

1.89

画像

1.86

Activations Density 0.101%