© 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. 31-GEMMASCOPE-2-RES-16K
    7. 5465
    Prev
    Next
    INDEX
    Explanations

    only

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

    The neuron strongly activates on numeric tokens (e.g. years, IDs, amounts, percentages), i.e. any standalone sequence of digits.

    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_31_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
    Defaults
    0.60
    ກະ
    0.58
     দক্ষ
    0.55
    blob
    0.54
    ücher
    0.53
    emd
    0.52
    //=
    0.52
    uves
    0.52
    ToBe
    0.51
    Default
    0.50
    POSITIVE LOGITS
    蝌
    0.56
    ワン
    0.55
    月的
    0.54
    只
    0.54
    न्
    0.53
    មុ
    0.53
    ൻ
    0.53
     Baillargeon
    0.53
    وانات
    0.52
    nań
    0.52
    Activations Density 0.000%

    No Known Activations