INDEX

Explanations

information released by

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

-1.41

Välisling

-1.33

-1.31

БК

-1.27

 ۱۵

-1.27

 với

-1.26

 recibe

-1.24

something

-1.23

 ۱۰

-1.22

 ۱۲

-1.22

POSITIVE LOGITS

 this

2.63

 that

2.08

 from

1.90

 which

1.81

or

1.63

of

1.57

 only

1.46

 other

1.46

 these

1.45

by

1.41

Activations Density 0.064%