INDEX

Explanations

maintaining appropriate

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

iettivo

-0.84

mbers

-0.82

 Ceramby

-0.81

Kyr

-0.76

识

-0.75

 Walden

-0.73

 obicei

-0.71

erequisites

-0.71

lī

-0.71

 Saxe

-0.70

POSITIVE LOGITS

 goog

1.81

goog

1.40

 Google

1.16

 Jersey

1.09

Google

1.05

 Rutgers

1.01

NJ

1.00

 jspb

0.99

 Gmail

0.94

nj

0.94

Activations Density 0.048%