INDEX

Explanations

different

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

×ľ×§

-0.08

æĲħ

-0.07

enco

-0.07

æĬĺ

-0.07

Reuse

-0.07

agar

-0.06

agina

-0.06

heritance

-0.06

[prop

-0.06

Ð±Ñĥ

-0.06

POSITIVE LOGITS

 differing

0.10

 whose

0.09

æĺ¯åħ·æľī

0.09

 possessing

0.09

 varying

0.09

ä½İèĩ³

0.09

 having

0.08

ä¸Ńåħ·æľī

0.08

åħ·æľī

0.08

whose

0.08

Activations Density 0.247%