INDEX

Explanations

fields requiring your info

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

but

-1.25

 sinds

-1.18

 aulas

-1.17

◾

-1.13

kui

-1.10

❎

-1.09

erdings

-1.09

 CERTAIN

-1.09

 xvii

-1.08

 kunder

-1.08

POSITIVE LOGITS

 your

2.03

you

1.36

 name

1.26

 space

1.11

 necessary

1.10

一旦

1.09

 number

1.07

 vorg

1.07

 affirmé

1.07

 either

1.06

Activations Density 0.018%