INDEX

Explanations

phony propaganda

negative outcomes and manipulation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

溉

0.44

 дом

0.43

VPN

0.42

 аллер

0.42

 요청

0.41

ुलेंस

0.41

ella

0.41

 их

0.40

陪伴

0.39

 आजही

0.39

POSITIVE LOGITS

 propagand

0.61

 propaganda

0.54

 phony

0.51

 mutil

0.50

mao

0.50

 misery

0.47

 mercantil

0.46

 insult

0.46

リッジ

0.45

 মিথ্যা

0.45

Activations Density 0.001%