© 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. 3762
    Prev
    Next
    INDEX
    Explanations

    publication/retrieval indicators

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

    The neuron is primarily detecting numeric tokens (years, page or volume numbers, retrieval dates, etc.) in reference and citation entries.

    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
     eventful
    0.63
     courageous
    0.61
     incontro
    0.60
     irresist
    0.59
    meldung
    0.59
    ッピング
    0.59
     injurious
    0.59
    隕
    0.59
     Botan
    0.59
     Suss
    0.59
    POSITIVE LOGITS
     Consultado
    0.82
     Ginkuhà
    0.77
     مؤرشف
    0.75
     Diakses
    0.67
     ശേഖ
    0.66
     тря
    0.64
     Viitattu
    0.64
     สืบค้นเมื่อ
    0.63
    brows
    0.62
    archiviato
    0.61
    Activations Density 0.029%

    No Known Activations