INDEX

Explanations

risk and preventative measures

common phrases and structures

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

MARINE

0.47

 zakup

0.47

 MARINE

0.47

 мор

0.46

ανα

0.45

🦈

0.45

ႃ

0.44

 stok

0.44

 tuesday

0.44

🌊

0.44

POSITIVE LOGITS

 Invasive

0.48

飕

0.48

相手

0.47

 Resistant

0.46

oids

0.46

 invasive

0.45

kens

0.44

 profitable

0.43

きましょう

0.42

 보다

0.42

Activations Density 0.000%