INDEX

Explanations

slash followed by acronyms

the forward slash character separating organizational or institutional acronyms and abbreviations.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 they

-2.28

-1.85

are

-1.84

it

-1.50

 rumahnya

-1.48

 Hinblick

-1.47

athons

-1.47

ICALS

-1.46

 weren

-1.45

POSITIVE LOGITS

 και

1.87

嵛

1.76

larınız

1.70

ropshire

1.66

to

1.63

และ

1.62

 এবং

1.58

⚌

1.57

ようになる

1.53

噠

1.51

Activations Density 0.042%