INDEX

Explanations

useful, interest, value, relevant seeking

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

The neuron fires on words that express a judgment of usefulness or interest—i.e. subjective evaluations like “useful,” “interesting,” or “value.”

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

language indicating relevance or value, such as statements that something is of interest, useful, or valuable.

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

the phrase "of interest" in technical or formal contexts.

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

각

-1.14

 each

-1.14



-1.13

ǀ

-1.10

 FontWeight

-1.09

🐟

-1.09

США

-1.09

 número

-1.08

Each

-1.06

-1.05

POSITIVE LOGITS

 enough

1.32

 этом

1.31

用到

1.26

 strona

1.26

 メニュー

1.19

犼

1.17

 gleaned

1.15

 drzwi

1.14

 wtedy

1.14

 around

1.14

Activations Density 0.140%

useful, interest, value, relevant seeking

The neuron fires on words that express a judgment of usefulness or interest—i.e. subjective evaluations like “useful,” “interesting,” or “value.”

language indicating relevance or value, such as statements that something is of interest, useful, or valuable.

the phrase "of interest" in technical or formal contexts.

No Comments

No Known Activations

useful, interest, value, relevant seeking

The neuron fires on words that express a judgment of usefulness or interest—i.e. subjective evaluations like “useful,” “interesting,” or “value.”

language indicating relevance or value, such as statements that something is of interest, useful, or valuable.

the phrase "of interest" in technical or formal contexts.

No Comments

No Known Activations