© 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. 2175
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on “/” tokens, especially when used in numeric or citation-like contexts (e.g. volume/issue, year/page separators).

    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
     јуну
    0.88
    ಂತಹ
    0.84
    ྨ
    0.83
    ersen
    0.79
    '%(
    0.77
     continueRoutine
    0.75
    症状
    0.74
    吵
    0.73
    ер
    0.73
     insolvent
    0.71
    POSITIVE LOGITS
    coordinate
    0.80
    delegate
    0.80
    deleg
    0.80
    trans
    0.79
    kn
    0.78
    data
    0.77
    tele
    0.75
    transport
    0.74
    finance
    0.73
    to
    0.72
    Activations Density 0.000%

    No Known Activations