INDEX

Explanations

is or whether questions

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ;)↵↵

-0.13

lol

-0.13

@_

-0.13

 ^↵

-0.12

^^

-0.12

yay

-0.12

;)

-0.12

Sadly

-0.12

 haha

-0.11

bu

-0.11

POSITIVE LOGITS

 darn

0.10

 darling

0.09

 Honey

0.09

 fellows

0.09

 collo

0.09

 dear

0.09

 Debbie

0.08

Gee

0.08

 Dick

0.08

gee

0.08

Activations Density 0.095%