INDEX

Explanations

if statements in code

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

to

-1.45

に行ってきました

-1.38

blusa

-1.38

 those

-1.36

perasi

-1.28

隴

-1.27

ﺢ

-1.27

 Vergrößern

-1.25

 odstran

-1.23

って何

-1.23

POSITIVE LOGITS

 REALLY

1.59

“

1.44

1.38

So

1.37

や

1.34

 وليس

1.25

 THERE

1.25

more

1.24

Activations Density 0.009%