INDEX

Explanations

burning

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

burns

-0.93

rungsseite

-0.92

 burns

-0.82

burn

-0.78

 burn

-0.73

 estekak

-0.71

tanleria

-0.69

Burns

-0.69

WireFormatLite

-0.69

 للمعارف

-0.69

POSITIVE LOGITS

up

0.79

off

0.60

Up

0.50

 away

0.47

 Theron

0.45

the

0.42

OFF

0.42

 forth

0.42

 Parac

0.41

Off

0.41

Activations Density 0.011%