INDEX

Explanations

seeing and perceiving

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

of

-2.11

if

-1.52

you

-1.40



-1.29

use

-1.28

on

-1.25

になるので

-1.25

 articol

-1.24

 wojsk

-1.24

of

-1.21

POSITIVE LOGITS

?,

1.29

龊

1.28

omás

1.25

 insumos

1.24

 alebo

1.24

 atau

1.24

 !...

1.22

AZIONI

1.21

 distribuir

1.18

 estabiliz

1.18

Activations Density 0.035%