INDEX

Explanations

Internet Protocol

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

the word "Internet" (especially when capitalized and used in technical or formal contexts).

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

The neuron strongly activates on occurrences of the term “Internet” (and closely associated protocol jargon like “Protocol”), effectively detecting explicit mentions of the Internet.

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

 моей

-2.06

鯷

-2.05

ヨーロ

-1.98

 моих

-1.95

</b>

-1.92

在我

-1.84

𖣔

-1.83

uver

-1.80

鉕

-1.80

トウ

-1.80

POSITIVE LOGITS

2.42

1.95

 Internet

1.85

 There

1.84

1.82

ÇÃO

1.80

it

1.79

圣节

1.78

there

1.76

博多

1.74

Activations Density 0.030%

Internet Protocol

the word "Internet" (especially when capitalized and used in technical or formal contexts).

The neuron strongly activates on occurrences of the term “Internet” (and closely associated protocol jargon like “Protocol”), effectively detecting explicit mentions of the Internet.

No Comments

No Known Activations

Internet Protocol

the word "Internet" (especially when capitalized and used in technical or formal contexts).

The neuron strongly activates on occurrences of the term “Internet” (and closely associated protocol jargon like “Protocol”), effectively detecting explicit mentions of the Internet.

No Comments

No Known Activations