INDEX

Explanations

church titles and organizations

The neuron activates on first-person, self-referential language—pronouns and verbs framing “I,” “my,” “me,” “myself,” etc.

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

if

-1.66

 туры

-1.59

𝐉

-1.58

 อาหาร

-1.56

 maravilla

-1.55

大な

-1.48

摂取

-1.46

🍗

-1.45

といふ

-1.45

any

-1.45

POSITIVE LOGITS

 zudem

1.62

1.59

 塗装

1.59

↵↵↵↵↵↵↵↵↵↵↵↵↵

1.53

웡

1.52

<bos>

1.52

</tbody>

1.49

 conteúdo

1.48

 próprio

1.47

Activations Density 0.005%