INDEX

Explanations

punctuation followed by words

np_acts-logits-general · gemini-2.5-flash-lite

This neuron spotlights editorial or expressive asides—parenthetical or bracketed interjections and markers like [sic], (gasp!), grin, (a hem), etc.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

parenthetical or bracketed asides and editorial annotations, especially the closing punctuation of such insertions.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

closing brackets or parentheses following interjections, editorial notes, or asides.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

長男

-1.62

やりたい

-1.48

蒐

-1.46

It

-1.45

家で

-1.41

どういう

-1.39

 allgemeine

-1.38

 simply

-1.37

かなぁ

-1.36

頂いた

-1.35

POSITIVE LOGITS

>−</

1.44

$=\

1.43

circledR

1.35

\%

1.34

数の

1.34

ました

1.32

 vitale

1.30

 ſever

1.30

 occidentale

1.30

０

1.30

Activations Density 0.007%

punctuation followed by words

This neuron spotlights editorial or expressive asides—parenthetical or bracketed interjections and markers like [sic], (gasp!), *grin*, (a hem), etc.

parenthetical or bracketed asides and editorial annotations, especially the closing punctuation of such insertions.

closing brackets or parentheses following interjections, editorial notes, or asides.

No Comments

No Known Activations

This neuron spotlights editorial or expressive asides—parenthetical or bracketed interjections and markers like [sic], (gasp!), grin, (a hem), etc.