INDEX

Explanations

actions followed by of

generation of prohibited content

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

They

0.43

 pracę

0.42

それを

0.42

 נישט

0.41

'.'

0.41

بر

0.40

Converting

0.40

ลูก

0.40

 swoim

0.40

িত

0.39

POSITIVE LOGITS

of

1.22

ของการ

1.05

 ofthe

1.02

 của

0.98

 των

0.89

នៃ

0.88

 של

0.86

 của

0.86

ของ

0.76

 τῶν

0.73

Activations Density 0.186%