INDEX

Explanations

production and consequences

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

ه

1.37

тились

1.31

1.30

問題

1.25

bl

1.24

就这样

1.24

Seeing

1.23

涵

1.22

AR

1.22

 coton

1.21

POSITIVE LOGITS

weisen

1.39

 maternal

1.38

ვე

1.33

거운

1.22

 superoxide

1.21

 anvil

1.20

 determining

1.20

արկ

1.19

 dị

1.17

erman

1.17

Activations Density 0.000%