INDEX

Explanations

form, version, adaptation, or conversion

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

↵↵↵

-1.40

↵

-1.36

 wich

-1.26

by

-1.11

 which

-1.10

 fand

-1.06

 verdi

-1.06

re

-1.06

嗆

-1.04

 जो

-1.03

POSITIVE LOGITS

 everything

1.46

tería

1.46

亊

1.35

 Ganzen

1.31

TabItem

1.27

omyia

1.23

 жовтня

1.23

 organisiert

1.23

 this

1.21

izamos

1.20

Activations Density 0.133%