© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 8505
    Prev
    Next
    INDEX
    Explanations

    domain, DNS, European regions

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

    The neuron strongly activates on promotional product-or service-marketing copy—i.e. the kind of text framing a brand or offering (e.g. “dev as a home for developers,” “Web Hosting services…”) rather than ordinary prose.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    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
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MaxIntensity
    0.59
     thermos
    0.55
     thermoplastics
    0.55
     intr
    0.54
     legislate
    0.53
    궤
    0.53
     paw
    0.52
     device
    0.52
    메이션
    0.52
     paws
    0.52
    POSITIVE LOGITS
    域名
    1.59
     Domain
    1.56
     Domains
    1.55
     domain
    1.54
    Domain
    1.53
     domains
    1.51
    domains
    1.48
     DNS
    1.48
    Domains
    1.48
    domain
    1.47
    Activations Density 0.359%

    No Known Activations