© 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. 3111
    Prev
    Next
    INDEX
    Explanations

    business, links, reports

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

    This neuron fires on the main content word that begins a title or heading in a citation (the first significant, often capitalized, word of an article or chapter title).

    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
     DataTable
    0.68
     */
    0.65
     func
    0.64
     expression
    0.64
     formData
    0.64
     static
    0.63
     Expression
    0.63
     LinearLayout
    0.62
     값이
    0.62
     ViewData
    0.62
    POSITIVE LOGITS
    bringing
    0.73
    combat
    0.71
    honor
    0.70
    covid
    0.70
    鉑
    0.69
    profit
    0.69
    dont
    0.68
    helping
    0.68
    for
    0.68
    太平洋
    0.67
    Activations Density 0.005%

    No Known Activations