INDEX

Explanations

**

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

The neuron strongly responds to named entities—especially multi‐word proper names such as brands, products, places, or people.

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

periods at the end of sentences.

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

key content nouns—especially abstract or category-defining terms often used in headings, lists, or formal/technical contexts.

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

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 அனைத்து

0.56

 nessuna

0.55

 Objekte

0.55

 एस

0.54

 ...)

0.54

 مثلا

0.50

⼤

0.50

 иллюстрация

0.49

ilustración

0.49

ស

0.48

POSITIVE LOGITS

<h4>

0.70

さて

0.68

<h1>

0.64

<h3>

0.64

こんにちは

0.62

<h2>

0.61

 こんにちは

0.61

Nowadays

0.61

。【

0.58

 गौरतलब

0.58

Activations Density 0.822%

**

The neuron strongly responds to named entities—especially multi‐word proper names such as brands, products, places, or people.

periods at the end of sentences.

key content nouns—especially abstract or category-defining terms often used in headings, lists, or formal/technical contexts.

No Comments

No Known Activations