© 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. 31-GEMMASCOPE-2-RES-16K
    7. 12032
    Prev
    Next
    INDEX
    Explanations

    after output

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

    The neuron fires strongly on words immediately adjacent to explicit numeric mentions (years, dates, counts, or other raw numbers), marking tokens that introduce or qualify a number.

    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
    ังก์
    0.71
    ic
    0.66
     wymaga
    0.65
    أي
    0.64
     secretário
    0.64
     चुंबकीय
    0.63
    शन
    0.63
    geojson
    0.61
    يو
    0.61
    ições
    0.61
    POSITIVE LOGITS
     школова
    0.71
    टकों
    0.70
     अवस्थ
    0.69
    嗒
    0.68
     шкі
    0.66
     ناح
    0.64
    ემი
    0.64
     Nunca
    0.64
     obses
    0.64
    耷
    0.64
    Activations Density 0.000%

    No Known Activations