INDEX

Explanations

URLs and web addresses

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

»)

-0.76

visation

-0.74

ficient

-0.73

 columnist

-0.72

expressing

-0.71

olate

-0.70

 alternatively

-0.69

"//

-0.69

Dismiss

-0.68

convertTo

-0.68

POSITIVE LOGITS

....

1.25

 ....

1.24

.....

1.19

)....

1.18

...."

1.13

......

1.06

...@

1.05

 .....

1.01

 ......

1.01

"....

0.96

Activations Density 0.237%