INDEX

Explanations

specifies how | specifies the | intonation contour | incredibly | incredulous | imploring

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

</b>

-3.45

-2.59

');

-2.44

砹

-2.28

 แต่

-2.27

ではまた

-2.27

 odpowiedzi

-2.23

Ārējās

-2.19

 متنوع

-2.17

 alters

-2.16

POSITIVE LOGITS

</h3>

3.66

ed

2.52

坶

2.45

 itſelf

2.31

if

2.20

áním

2.19

 gleichen

2.14

魮

2.08

getting

2.06

 “[

2.05

Activations Density 0.003%