INDEX

Explanations

reunited or proximity to beloved

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

ikers

-0.10

iasi

-0.10

 rival

-0.10

 pred

-0.10

 ê²

-0.09

 olarak

-0.09

 approving

-0.09

Zy

-0.09

 fascination

-0.09

oney

-0.09

POSITIVE LOGITS

 reunited

0.26

 reunion

0.25

 reun

0.23

 clos

0.16

 beloved

0.16

 proximity

0.16

 touch

0.15

 touching

0.15

 reconnect

0.14

 Touch

0.13

Activations Density 0.061%