INDEX

Explanations

Wikipedia content and links

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

𝕣

1.44

ээ

1.42

 ماذا

1.41

 sirens

1.39

*}\

1.37

ัฐ

1.34

front

1.31

গি

1.29

nö

1.26

褰

1.26

POSITIVE LOGITS

zil

2.01

zare

1.74

UNIQUE

1.73

nych

1.68

zele

1.66

टाइ

1.65

omal

1.62

зера

1.57

ন

1.56

oung

1.55

Activations Density 0.001%