INDEX

Explanations

a followed by descriptive word

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

it

-2.33

-1.32

ارتمان

-1.29

举动

-1.27

its

-1.27

 Institutionen

-1.27

してみた

-1.23

 while

-1.20

riwal

-1.20

and

-1.16

POSITIVE LOGITS

呿

1.55

smh

1.39

𓃵

1.30

investissement

1.29

oría

1.29

 tantum

1.26

 यदि

1.23

emplacement

1.23

duled

1.21

prende

1.20

Activations Density 0.022%