INDEX

Explanations

protecting against harm

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

箆

-1.45

內容簡介

-1.37

ܜ

-1.31

of

-1.30

دود

-1.29

ется

-1.28

 spettacolo

-1.27

 alerg

-1.26

 проблема

-1.25

 prate

-1.25

POSITIVE LOGITS

 nieco

1.60

1.57

ização

1.46

 sieben

1.45

馥

1.42

 tych

1.41

ICOLON

1.40

 haven

1.39

ícias

1.37

低于

1.36

Activations Density 0.064%