© 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. 3429
    Prev
    Next
    INDEX
    Explanations

    displayed_word substitution

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

    The neuron primarily fires on floating-point numbers (tokens containing digits with a decimal point).

    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
    objClass
    0.83
    <0x91>
    0.82
     freshly
    0.77
    ostí
    0.76
     विनो
    0.75
    🛖
    0.75
     হাস্য
    0.75
     enormes
    0.74
    왕
    0.74
     práv
    0.74
    POSITIVE LOGITS
    s
    0.78
    [
    0.76
    :
    0.69
    Yup
    0.68
     scarring
    0.67
    CUT
    0.65
    ler
    0.64
    Mumbai
    0.64
    siz
    0.64
    loat
    0.63
    Activations Density 0.001%

    No Known Activations