INDEX

Explanations

names of people and titles

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

The main thing this neuron does is identify personal names (proper nouns referring to people).

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

proper names of people, particularly when they appear in narrative contexts.

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

mentions of animate actors, especially human individuals—proper names, roles/titles, kinship terms, and generic person references.

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

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/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

 utilizz

-1.13

に入れる

-1.13

帰って

-1.13

に入れ

-1.13

ところに

-1.12

-1.11

様な

-1.09

すっ

-1.09

際は

-1.09

POSITIVE LOGITS

籐

1.16

Sabes

1.15

がある

1.15

umers

1.12

 sofá

1.10

moiselle

1.09

Сейчас

1.09

ÓD

1.06

Todavía

1.04

thenia

1.04

Activations Density 0.004%

names of people and titles

The main thing this neuron does is identify personal names (proper nouns referring to people).

proper names of people, particularly when they appear in narrative contexts.

mentions of animate actors, especially human individuals—proper names, roles/titles, kinship terms, and generic person references.

No Comments

No Known Activations

names of people and titles

The main thing this neuron does is identify personal names (proper nouns referring to people).

proper names of people, particularly when they appear in narrative contexts.

mentions of animate actors, especially human individuals—proper names, roles/titles, kinship terms, and generic person references.

No Comments

No Known Activations