INDEX

Explanations

cover art and interviews

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 cover

-1.50

Cover

-1.16

cover

-1.16

 covers

-1.13

 Cover

-1.08

 COVER

-1.05

封面

-1.04

COVER

-1.02

 logo

-0.93

 prowad

-0.91

POSITIVE LOGITS

 image

1.20

 changed

0.90

 graces

0.89

 प्य

0.84

 chosen

0.84

image

0.82

oraz

0.81

 progre

0.80

Chosen

0.80

 Utara

0.80

Activations Density 0.026%

No Known Activations