INDEX

Explanations

conduct code integrity entrepreneur steal

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

র্ষ

0.53

 বিস্ম

0.49

在

0.48

到底

0.48

ِيم

0.46

ក្នុងការ

0.45

)。

0.44

なのに

0.44

亟

0.43

]。

0.43

POSITIVE LOGITS

 Ethel

0.50

an

0.50

 Sabre

0.48

 Phyllis

0.46

 pess

0.45

CFC

0.44

 Piers

0.44

UBS

0.43

 Elsie

0.42

 Gladys

0.42

Activations Density 0.001%