INDEX

Explanations

cyber security and crime

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

the term "cyber" and related cybersecurity terminology.

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

The neuron strongly activates on occurrences of “cyber,” “security,” or “cybersecurity” (and related compound terms).

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

references to cybersecurity and related threats, operations, policies, and organizations.

oai_token-act-pair · gpt-5 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

-2.45

↵

-2.36

</h2>

-1.97

</h3>

-1.77

瀦

-1.71

</em>

-1.59

 landmark

-1.58

没了

-1.52

 quell

-1.52

 mám

-1.50

POSITIVE LOGITS

 Inſ

1.95

 Reſ

1.90

iembre

1.80

 Perſ

1.77

borderSide

1.73

secretary

1.69

 пребы

1.67

ATEGY

1.66

ETERS

1.64

 短发

1.64

Activations Density 0.011%

cyber security and crime

the term "cyber" and related cybersecurity terminology.

The neuron strongly activates on occurrences of “cyber,” “security,” or “cybersecurity” (and related compound terms).

references to cybersecurity and related threats, operations, policies, and organizations.

No Comments

No Known Activations

cyber security and crime

the term "cyber" and related cybersecurity terminology.

The neuron strongly activates on occurrences of “cyber,” “security,” or “cybersecurity” (and related compound terms).

references to cybersecurity and related threats, operations, policies, and organizations.

No Comments

No Known Activations