© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9668
    Prev
    Next
    INDEX
    Explanations

    authentication, tax, models, texture

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

    The neuron lights up almost exclusively on words that come from document “headings” or named‐entity–style phrases (proper names, section titles, question headings, etc.).

    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
    s
    0.84
    BOUND
    0.80
    hasOwnProperty
    0.79
    i
    0.79
    y
    0.75
    f
    0.75
     JOB
    0.73
    sizer
    0.73
    wander
    0.73
    ためには
    0.73
    POSITIVE LOGITS
     separators
    0.79
     rä
    0.78
    两天
    0.77
    多人
    0.75
     ذخ
    0.75
     kiire
    0.74
    违
    0.74
    ץ
    0.72
    िन
    0.71
     นี้
    0.71
    Activations Density 0.000%

    No Known Activations