INDEX

Explanations

nickname or alias followed by description

The neuron detects name‐introduction contexts—phrasing that signals someone’s given name, alias, or what they’re “called” or “known as,” often marked by commas, quotes, or parentheses.

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

 that

-1.50

</em>

-1.45

</h4>

-1.42

 ważne

-1.37

↵↵↵↵

-1.34

you

-1.34

↵↵↵↵↵

-1.33

↵↵↵

-1.31

 occur

-1.28

 deste

-1.27

POSITIVE LOGITS

〉

1.36

涠

1.34

cjs

1.34

最初に

1.28

 Перейти

1.22

]}

1.19

}},

1.17

’

1.16

 výrob

1.15

*<

1.15

Activations Density 0.047%