INDEX

Explanations

and/or, that/while/than

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

it

-1.71

to

-1.59

战胜

-1.50

却被

-1.48

agic

-1.46

帳に追加

-1.46

 सत्यापित

-1.45

却又

-1.45

for

-1.45

However

-1.44

POSITIVE LOGITS

1.87

ÜN

1.48

 dibuat

1.42

ֺ

1.41

„

1.39

Harp

1.38

of

1.36

 katholischen

1.32

 vært

1.29

Activations Density 0.128%