INDEX

Explanations

certain followed by nouns

The neuron specifically flags occurrences of the word “certain.”

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

of

-2.36

 інших

-2.09

 genoux

-2.06

 料金

-1.98

 chimiques

-1.96

 której

-1.89

 partenza

-1.89

ਃ

-1.86

if

-1.86

 notizie

-1.83

POSITIVE LOGITS

1.98

が多い

1.93

 pecu

1.73

 augus

1.73

 inger

1.73

 amount

1.72

 delin

1.69

 infantil

1.69

—

1.67

 pedag

1.61

Activations Density 0.012%