INDEX

Explanations

words and short phrases that indicate progress or advancement in a field or study

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Audiodateien

-0.79

 myſelf

-0.73

 duquel

-0.69

 purpoſe

-0.68

Viki

-0.66

 faſt

-0.63

ApiModelProperty

-0.63

 itſelf

-0.63

inaldi

-0.62

shadowRadius

-0.62

POSITIVE LOGITS

<bos>

0.56

Jeografia

0.55

SequentialGroup

0.54

contentLoaded

0.47

 CURIAM

0.45

 Taktlose

0.45

ArrowToggle

0.44

CPtr

0.44

 Sixt

0.43

LookAnd

0.42

Activations Density 0.450%