© 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. 2084
    Prev
    Next
    INDEX
    Explanations

    two or more items

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

    The neuron fires strongly on numeric tokens—especially decimal‐point numbers or measurements.

    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
     này
    0.70
    你
    0.69
    我
    0.64
    社会
    0.62
     nào
    0.60
    écoute
    0.57
    з
    0.57
    с
    0.55
    ：
    0.55
    没有
    0.55
    POSITIVE LOGITS
     ayrıca
    0.68
     ardından
    0.64
     two
    0.63
     twofold
    0.62
     three
    0.61
     tiež
    0.61
     valamint
    0.59
     zudem
    0.59
     four
    0.58
     also
    0.57
    Activations Density 0.200%

    No Known Activations