INDEX

Explanations

drug dealing and trafficking

The main thing this neuron does is detect mentions of drugs and drug-related terms (e.g., drug, drugs, trafficking, dealers, abuse).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-1.72

<tr>

-1.59

 friv

-1.54

-%

-1.46

ers

-1.45

the

-1.44

::

-1.43

ns

-1.41

POSITIVE LOGITS

oyant

1.59

drug

1.56

のよう

1.48

trale

1.46

 vollständ

1.40

戬

1.39

abilité

1.39

 Таким

1.36

OTROS

1.33

怔

1.32

Activations Density 0.061%