INDEX

Explanations

true meaning or significance

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

 verschieden

-1.45

 aussi

-1.39

 auch

-1.38



-1.28

٨

-1.27

 ако

-1.23

縢

-1.22

⢅

-1.22

csol

-1.21

and

-1.20

POSITIVE LOGITS

↵

1.95

 then

1.38

jestel

1.36

लेट

1.24

 they

1.23

UDIO

1.22

1.21

 should

1.20

imų

1.20

 Suiza

1.16

Activations Density 0.037%