INDEX

Explanations

job titles followed by roles or descriptions

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

the

-1.63

 decidi

-1.34

 zobaczyć

-1.33

 fameux

-1.31

値下げ

-1.26

 nationaux

-1.25

 facilité

-1.24

 スリーブ

-1.24

 männis

-1.22

 specchio

-1.20

POSITIVE LOGITS

in

2.33

at

1.93

1.82

 without

1.63

 with

1.49

 from

1.48

즘

1.40

or

1.39

 Перейти

1.37

for

1.34

Activations Density 0.060%