INDEX

Explanations

actions, remorse, and change

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

勿論

-1.41

-1.35

卫生间

-1.30

除此

-1.27

 atractiva

-1.23

 reconocida

-1.20

 megfe

-1.20

但她

-1.18

of

-1.18

足の

-1.17

POSITIVE LOGITS

.”

1.48

ların

1.36

mería

1.34

那些

1.30

 behavior

1.29

 Réalisation

1.28

Ꮗ

1.27

，“

1.24

ոյ

1.23

↵

1.23

Activations Density 0.019%