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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6516
    Prev
    Next
    INDEX
    Explanations

    data types and structures

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

    This neuron detects labels or tokens marking structured personal or contact information fields (e.g. email, phone number, address, race, residence, date/place of birth).

    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_24_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
    tanh
    0.63
    Weyl
    0.59
     champignons
    0.59
    Phosph
    0.58
    vané
    0.58
    澼
    0.58
    মিশ্র
    0.57
    catalyzed
    0.57
     सुगंध
    0.57
    穵
    0.57
    POSITIVE LOGITS
     startDate
    0.82
     эмоциона
    0.81
     username
    0.77
     firstName
    0.77
     date
    0.75
     createdAt
    0.75
     timestamp
    0.73
     varchar
    0.73
     финансо
    0.73
     ethnicity
    0.71
    Activations Density 0.073%

    No Known Activations