© 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. 10685
    Prev
    Next
    INDEX
    Explanations

    copy and paste

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

    This neuron fires on tokens with unusually large embedding magnitudes—essentially the very high‐frequency/common tokens (e.g. “the,” “both,” “fee,” “mpg,” “titer”).

    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
    ফ
    0.75
    通过
    0.75
    pand
    0.70
     Int
    0.68
    在这
    0.68
    ures
    0.68
    cion
    0.67
    brane
    0.67
     Mechan
    0.66
    Dh
    0.66
    POSITIVE LOGITS
    ㄛ
    0.94
     Jehova
    0.89
    و
    0.84
     Compaq
    0.82
     Сергея
    0.82
     Toshiba
    0.81
    <0x07>
    0.80
     MSNBC
    0.80
     ちゃう
    0.79
     Marcello
    0.78
    Activations Density 0.000%

    No Known Activations