INDEX

Explanations

but followed by contrast

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

ing

-1.25

няя

-1.23

 where

-1.21

on

-1.20

 généralement

-1.18

Beneficios

-1.16

ian

-1.16

闋

-1.15

year

-1.15

</b>

-1.15

POSITIVE LOGITS

 серд

1.41

 stadig

1.37

 informací

1.34

rimos

1.31

𓏸

1.30

stoy

1.28

 groei

1.27

讠

1.24

吀

1.24

 للع

1.24

Activations Density 0.069%