© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 12885
    Prev
    Next
    INDEX
    Explanations

    notes and digit/numeric sequences

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

    This neuron is highly selective for the token “Day” (as in holiday names like “Memorial Day”).

    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
    ič
    0.68
    ių
    0.66
     поду
    0.61
    ()){
    0.61
    ACH
    0.60
    Rf
    0.59
     আদাল
    0.59
    )&=
    0.59
    Cz
    0.59
     речо
    0.58
    POSITIVE LOGITS
    йде
    0.81
    <0xA0>
    0.79
     freshest
    0.79
    <0xAF>
    0.77
    mba
    0.76
    в
    0.76
     dearth
    0.75
    т
    0.75
    hava
    0.75
    с
    0.75
    Activations Density 0.000%

    No Known Activations