INDEX

Explanations

Mala, Rosario, leprosy, tables

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

words related to specific proper nouns and professional terminology.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron fires most strongly on standalone proper‐name tokens or title/heading words (e.g. names of societies, places, works, headings), i.e. tokens marking article subjects or section titles.

oai_token-act-pair · o4-mini 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

 王子

-1.73

푙

-1.66

 天空

-1.66

 航空

-1.65

</b>

-1.62

nodata

-1.62

 خدم

-1.59

caution

-1.58

 ajudar

-1.55

 pion

-1.55

POSITIVE LOGITS

</strong>

2.22

“

2.19

1.94

 worin

1.83

</h3>

1.82

 Rosary

1.80

of

1.73

("

1.72

ה

1.70

1.69

Activations Density 0.001%

Mala, Rosario, leprosy, tables

words related to specific proper nouns and professional terminology.

The neuron fires most strongly on standalone proper‐name tokens or title/heading words (e.g. names of societies, places, works, headings), i.e. tokens marking article subjects or section titles.

No Comments

No Known Activations

Mala, Rosario, leprosy, tables

words related to specific proper nouns and professional terminology.

The neuron fires most strongly on standalone proper‐name tokens or title/heading words (e.g. names of societies, places, works, headings), i.e. tokens marking article subjects or section titles.

No Comments

No Known Activations