© 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. 6656
    Prev
    Next
    INDEX
    Explanations

    Billion years ago

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

    The neuron strongly activates on standalone numeric tokens or digit sequences (e.g. numbers, measurements, or counts).

    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
    和服务
    0.77
    ///
    0.76
    نائي
    0.74
    //////
    0.74
    ////
    0.73
     દરમ
    0.72
    یز
    0.71
    Locale
    0.71
    ين
    0.70
    ترنت
    0.70
    POSITIVE LOGITS
    ﺏ
    0.91
     сказать
    0.89
     ног
    0.83
     Stunden
    0.82
     schlecht
    0.82
     cuja
    0.82
     эмне
    0.81
     требуют
    0.81
     поэтому
    0.80
     sprang
    0.80
    Activations Density 0.001%

    No Known Activations