INDEX

Explanations

industrial

understanding perpetrator behaviour

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 đặt

0.58

 рука

0.52

ंपरिक

0.50

 phần

0.50

硙

0.49

 цик

0.49

 Convenient

0.48

 část

0.48

部分的

0.48

也没有

0.48

POSITIVE LOGITS

ולוג

0.54

р

0.48

정

0.48

Gang

0.47

ü

0.46

Gary

0.46

industrial

0.44

Gert

0.43

옌

0.43

Industrial

0.43

Activations Density 0.000%