INDEX

Explanations

as a or as an followed by a role

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

 various

-1.44

 this

-1.37

 when

-1.25

if

-1.20

 suitable

-1.19

 leal

-1.18

it

-1.14

 each

-1.12

 conector

-1.12

 válvula

-1.11

POSITIVE LOGITS

 égard

1.45

 perpétu

1.27

also

1.23

Its

1.19

őség

1.18

always

1.18

骉

1.17

only

1.16

uming

1.16

 chargée

1.16

Activations Density 0.025%