INDEX

Explanations

assembly code and strings

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

 trzech

-1.49

those

-1.42

 vertes

-1.39

最后由

-1.38

쏙

-1.37

 シンプル

-1.34

他們

-1.32

ומי

-1.30

떄

-1.30

?,

-1.30

POSITIVE LOGITS

in

1.93

but

1.67

).

1.61

at

1.59

1.52

 while

1.48

 ....

1.44

 since

1.44

 from

1.41

Activations Density 0.007%